Вы находитесь на странице: 1из 35

A Review of Music and Emotion Studies: Approaches, Emotion Models, and Stimuli

Author(s): Tuomas Eerola and Jonna K. Vuoskoski


Source: Music Perception: An Interdisciplinary Journal, Vol. 30, No. 3 (February 2013), pp.
307-340
Published by: University of California Press
Stable URL: http://www.jstor.org/stable/10.1525/mp.2012.30.3.307 .
Accessed: 06/05/2013 08:34
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp

.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.

University of California Press is collaborating with JSTOR to digitize, preserve and extend access to Music
Perception: An Interdisciplinary Journal.

http://www.jstor.org

This content downloaded from 129.240.0.83 on Mon, 6 May 2013 08:34:26 AM


All use subject to JSTOR Terms and Conditions

Review of Music and Emotion Studies

307

A REVIEW OF MUSIC AND EMOTION STUDIES:


APPROACHES, EMOTION MODELS, AND STIMULI
T U O M A S E E R O L A & J O N N A K. V U O S K O S K I
University of Jyvaskyla, Jyvaskyla, Finland
THE FIELD OF MUSIC AND EMOTION RESEARCH HAS
grown rapidly and diversified during the last decade.
This has led to a certain degree of confusion and inconsistency between competing notions of emotions, data,
and results. The present review of 251 studies describes
the focus of prevalent research approaches, methods,
and models of emotion, and documents the types of
musical stimuli used over the past twenty years.
Although self-report approaches to emotions are the
most common way of dealing with music and emotions,
using multiple approaches is becoming increasingly
popular. A large majority (70%) of the studies employed
variants of the discrete or the dimensional emotion
models. A large proportion of stimuli rely on a relatively
modest amount of familiar classical examples. The evident shortcomings of these prevalent patterns in music
and emotion studies are highlighted, and concrete plans
of action for future studies are suggested.
Received: October 28, 2011, accepted April 17, 2012.
Key words: emotion, review, music, stimuli, theoretical

T HAS OFTEN BEEN SUGGESTED THAT THE

emotional effects of music are the most important


reason why people engage in musical activities (Juslin & Laukka, 2004; Sloboda & ONeill, 2001). The emotional power of music is the reason for its application in
areas as diverse as the gaming industry, film industry,
marketing, and music therapy, yet the scientific insights
into this phenomenon are far from complete or revealing. Contemporary research on music and emotion is
a popular topic within the fields of music cognition,
music psychology, and neuroscience of music, and
isby definitioninterdisciplinary. To date, a large
number of research approaches have been used to
explore why and how music has such a strong grip on
listenersregardless of differences in education,

Music Perception, V OLU M E 30, IS S U E 3,


R IG HTS RES ER V ED . PL EASE D IREC T AL L

PP .

307340,

IS S N

0730-7829,

personality, musical taste, or culture. The contents of


music that induces emotion have been extracted and
manipulated in numerous ways in order to understand
the cues that lead us to assign emotional labels to music.
And over the years, a large variety of musicfrom artificial melodies to live opera performanceshas been
used for this purpose. Because of the range of
approaches, models and settings used in music and
emotion studies, one of the most current and potent
criticisms is that all this research has not really resulted
in a set of coherent findings since large discrepancies
exist concerning exactly what is being studied (e.g., Juslin & Vastfjall, 2008). This is not entirely surprising
given that emotion research in general has acknowledged the challenges in defining emotions in a way that
would be acceptable to most researchers (Frijda, 2007;
Izard, 2007). Emotions in music may even be a more
complicated topic since we enter the realm of aesthetics
and so-called positive emotions, that are in contrast to
the survival-oriented emotions (such as fear and anger)
that have received far greater attention within mainstream psychology.
The most fundamental question addressed in music
and emotion studies is arguably, How does music
evoke emotions in listeners? This question can be broken down into separate areas of inquiry, such as: 1)
What are the putative emotions induced by music, and
which components contribute to these? 2) How are
emotions conveyed by music (e.g., elements of music,
lyrics, sound quality, etc.)? 3) What are the contributions of the situation (e.g., alone/in company, different
activities), listener attributes (e.g., expertise, personality), and the intention to regulate ones own affective
state? 4) Are the processes involved largely learned or
universal, and how are they related to other perceptual,
cognitive, and meaning-generation processes in our
minds? Most such pivotal issues branching out from
the fundamental question require multiple approaches
(e.g., cross-cultural, biological, developmental) in order
to be investigated fruitfully and convincingly. The
whole definition of the phenomenon itself is intertwined with the theoretical and operational notions of
different emotion models, which give rise to entirely
different assumptions of the processes involved (e.g.,

EL ECTR ON IC I SSN

1533-8312. 2013

B Y T HE R E G E NTS OF THE UN IV E RS IT Y OF CA LIF OR NIA A L L

REQ UEST S F OR PER MI SSION T O PHOT OCOPY O R REPR OD UCE AR T ICLE CON TEN T THR OUGH THE UN IVERS I TY OF C AL IFO RN IA PR E SS S

RI GHT S A ND PER MIS S IO NS W EBS I TE , H TT P :// W W W . U CPRE SS J OU RNA L S . COM / REPR INT INF O . A S P .

DOI: 10.1525/ M P .2012.30.3.307

This content downloaded from 129.240.0.83 on Mon, 6 May 2013 08:34:26 AM


All use subject to JSTOR Terms and Conditions

308

Tuomas Eerola & Jonna K. Vuoskoski

automatic and universal basic emotions versus highly


specific, aesthetic feelings of awe). Finally, the concept
of music itself isnaturallyalso a central building
block of the fundamental question, and it is worth asking what kinds of music have been used in the studies
exploring these core topics. By dissecting the fundamental question into these broad elements, we aim to
review the recent studies on music and emotion, and to
provide a summary of the approaches, emotion models,
and musical stimuli used in these studies. Summarizing
the past research efforts in a novel way may allow us to
highlight critical issues that have been previously overlooked. Do we take it for granted that basic emotions are
the most commonly studied emotions in the field of
music and emotions? How common is it to combine
self-reports with physiological measures and individual
variables such as personality traits? How is the issue of
emotion recognition vs. emotion induction distributed
across all the research approaches and emotion models
used? Do the stimuli used in emotion studies reflect the
everyday experiences and tastes of music listeners?
These are all examples of questions that are important
for the growing numbers of researchers coming into the
field, and may allow those already familiar with the field
to investigate more precisely and systematically the
ways in which music interacts with the emotional lives
of everyday listeners.
After the publication of Juslin and Slobodas (2001)
influential handbook on music and emotions, a number
of overviews have documented the state of the field.
These studies have been made from a wide range of
perspectives. Vastfjall (2002) qualitatively reviewed
mood induction procedures in a host of studies (including marketing studies) and suggested recommendations
for future studies. An influential meta-analysis by Juslin
and Laukka (2003) analyzed the communication accuracy of emotions via music and speech using strict
meta-analysis criteria. For the present work this is the
most directly relevant previous study, although their
meta-analysis was restricted to studies that employed
discrete emotions. One of the central findings of this
meta-analysis was to show the overall decoding accuracy of emotions. Our aim is to update the findings to
see whether researchers in the field have indeed still
almost exclusively studied discrete emotions (p. 776,
Juslin & Laukka, 2003). Instead of focusing on communication accuracy, we aim to provide an updated overview of the approaches, models, and stimuli employed
within the field. Juslin and Laukka also provided a partial summary of the stimuli used in the reviewed studies
(genres, instruments, paradigms for generating the
stimuli) and concluded that a wide range of musical

styles is used in music and emotion studies, classical


music being the most common. Although we intend to
characterize the stimulus materials in a more exhaustive
fashion, this previous work will also provide an interesting point of comparison. The third comprehensive
survey of music and emotion studies is the work by
Gabrielsson and Lindstrom (2001), which focused on
performance and structural features that are assumed to
contribute to emotional expression in music. Their
summary covered 32 empirical studiesstarting chronologically from Hevner (1936) through to Balkwill and
Thompson (1999)that examine the role of separate
musical factors (tempo, articulation, harmony, timbre,
etc.) in the expression of different emotions. As well as
these major summaries, a number of journal articles by
eminent scholars have made succinct digests of previous
research in the field, and various suggestions for further
research (e.g., Juslin & Sloboda, 2010; Juslin & Vastfjall,
2008; Koelsch, Fritz, Cramon, Muller, & Friederici,
2006; Peretz, 2010).
To conclude, the accumulated mass of music and
emotion studies has reached a critical point, in so far
that it has become increasingly diverse in terms of
research approaches, methods, and models, to the point
of losing consistency. There is the ever-growing sense
that studies thus far have produced data that are collectively confusing and internally inconsistent (Juslin
& Vastfjall, 2008, p. 574). This confusion is mainly
caused by variations in models, terminologies, and
underlying mechanisms. In the following pages, we will
first attempt to describe the dominant trends in music
and emotion studies, and then address the evident
shortcomings of these studies in detail.
Review Method
INCLUSION CRITERIA

To form an overview of the approaches, emotion models, and stimuli employed within the field of music and
emotions, we reviewed studies published since 1988
until February 2009 in refereed journals or conference
proceedings, thus spanning 20 years of studies. Despite
the scattered nature of music and emotion publication
channels (psychology, engineering, musicology, acoustics, and even medicine), we attempted to collect the
essential studies from PsycINFO and RILM using
suitable keywords (music, emotion, affect*, mood,
express*). We also used various other sources to supplement these studies. To constrain the number of studies we left out individual book chapters, non-English
texts, and unpublished dissertations. A total of 251
studies were annotated in terms of the (1) research

This content downloaded from 129.240.0.83 on Mon, 6 May 2013 08:34:26 AM


All use subject to JSTOR Terms and Conditions

Review of Music and Emotion Studies

approach, (2) emotion model, (3) locus of emotion, and


(4) stimulus details. Focusing in this way ensured that
the work already done by Juslin and Laukka (2003) was
minimally replicated and so our work would provide
a different, updated, and broader perspective to what the
popular choices made by researchers within the field are.
RESEARCH APPROACHES

During the review process, research approaches were


divided into seven broad categories: theoretical, selfreport, biological, clinical, developmental, music-analytic,
and individual and cultural differences. These categories
are not mutually exclusive as a study might employ both
biological and self-report approaches (to give an example of a common overlap). Examples of studies that
represent each category will be given later.
The theoretical approach consists of non-empirical
formulations of emotions within the domain of music,
including review studies. In other words, approaches
without empirical evidence to support them. It should
be noted that of course all empirical studies also use
theoretical assumptions in order to falsify or corroborate these concepts; however, for the purposes of this
review they have been classified under the type of
empirical approach adopted. This definition also means
that certain important theoretical contributions with
a major empirical part are subsumed under different
approaches. However, these few exceptions, namely, the
application of the brunswikian lens model to music and
emotion by Juslin (2000), proposal of the GERM model
by Juslin, Friberg, and Bresin (2001), and the GEMS
model by Zentner, Grandjean, and Scherer (2008) will
be covered and discussed in the summary sections. Also,
theoretical accounts may be more often documented in
monographs than in articles (e.g., Juslin & Sloboda,
2001; Kivy, 1990; Nussbaum, 2007), which render them
invisible for this review. The self-report category therefore contains all studies that utilize a paradigm where
a non-clinical sample of listeners or participants evaluates emotions either in music (expressed by music) or
induced by music using any type of self-report method
such as recognition tasks, Likert ratings, adjective lists,
free verbal reports, diaries, and also non-verbal
response formats such as assessing the similarity of two
examples. The biological approach is appropriate for
studies that employ any peripheral measure (e.g., skin
conductance, heart rate, muscle tension, respiration,
electrocardiogram, blood pressure, etc.) or a measure
of the central nervous system (e.g., EEG, MEG, fMRI,
PET). In neurological studies that involve pathologies, it
should be noted that the research approach is not only
neural, but also clinical.

309

The clinical approach means that these studies are


based on clinical populations (e.g., brain pathology or
a clinical group such as depressed patients). In spite of
these studies typically using a control group consisting
of a non-clinical population, they remain classified
under the clinical approach, as their reactions are being
noted only for their value as a control group. The term
developmental approach is applied when the participants of the study are children (typically 3-8 years of
age) and the main aim of the study is to examine developmental issues related to music and emotions. The
music analytic approach refers to studies in which the
actual musical features related to emotions are given an
emphasis by means of music analysis (acoustic, musicological, or based on ratings of musical features by
listeners), or direct manipulation of the particular musical features related to emotions. We do not consider
a study manipulating arousal by means of selecting different kinds of music for low and high arousal conditions as music analytic unless the differences in
collected emotion data are explained in terms of musical
differences between the stimuli. The final category of
approachindividual and cultural differences
includes studies relating to personality and expertise
as well as cross-cultural comparisons of emotional processing. Although the attention of cross-cultural comparisons is on intercultural differences rather than
interindividual differences, both share a similar interest
towards the generalizability of the findings of music and
emotion research across populations, and explore how
learning contributes to the recognition and induction
of emotions in the context of music (cf. Elfenbein &
Ambady, 2002). Application areas for music and
emotion research, such as consumer psychology and
purely medically oriented studies, were deemed to be
outside of the scope of this review. Similarly, studies
relying mainly on a computational approach to emotions were excluded in order to constrain the focus of
the review.
EMOTION MODELS

The theoretical models of emotion taken up by the


studies were divided into four classes: discrete, dimensional, miscellaneous, and music-specific. This division is
modified from the major theoretical traditions of emotion measurement (Barrett & Wager, 2006; Mauss &
Robison, 2009; Scherer, 2004) although it may not accurately distinguish the source (basic emotion, core affect,
music) and the structure (discrete or dimensional) of
emotions within the particular models, as discussed
later. The discrete model of emotion relates to the theory of basic emotions, which asserts that all emotions

This content downloaded from 129.240.0.83 on Mon, 6 May 2013 08:34:26 AM


All use subject to JSTOR Terms and Conditions

310

Tuomas Eerola & Jonna K. Vuoskoski

can be derived from a finite set of innate basic emotions.


These typically include fear, anger, disgust, sadness, and
happiness (Ekman, 1992; Panksepp, 1998), but may also
include emotions such as shame, embarrassment, contempt, and guilt (see Ortony & Turner, 1990). In this
review, studies within the discrete framework for
emotions include all variants of the basic emotions,
even if only a limited number of emotion categories
have been used (e.g., happy and sad). These are the
type of studies Juslin and Laukka reviewed in 2003
and their summary included the type of emotion
terms used and the number of emotions typically
included in the studies. For this reason, we will not
replicate these details here.
The dimensional model for emotions is the other
commonly used emotion model in psychology, and the
most famous example of this is the circumplex model
(Russell, 1980). This model represents emotions as
a mixture of two core dimensions, valence and arousal,
which are two continua that are orthogonally situated in
the affective space. This model has received support
in large-scale studies of self-reported emotions (e.g.,
Barrett & Russell, 1999), a cross-cultural comparison
(Russell, 1983), and psychometric studies (reviewed in
Posner, Russell, & Peterson, 2005). In this review, the
dimensional model for emotions includes all studies
that employ valence (or pleasant unpleasant),
arousal/activity, or other theoretically dimensional concepts (e.g., potency, tension arousal, etc.).
Miscellaneous models for emotion in music include
a motley collection of emotion concepts such as intensity, preference, similarity, tension, or some other construct closely linked to emotions in general. Although
the term tension was featured in two of our categories of
emotion model, deciding the appropriate one was fairly
straightforward as tension has been used in the musicspecific sense in several studies (Krumhansl, 1998;
McAdams et al., 2004) and as a dimensional construct
by only a few (e.g., Ilie & Thompson, 2006). Rather than
including a myriad of studies featuring preference or
liking in the context of music, we only included such
studies of preference to this review that contained an
explicit emotion task.
Both discrete and dimensional emotion models concentrate on a small set of evolutionary emotions that
have important functions when adapting the individual
to events that have material consequences for the individuals well-being. In contrast, music rarely has any such
effects on the individuals physical or psychological integrity. This issue has been taken up by work that has initially
compiled emotion lists and extracted the underlying
factors that are directly relevant for music (Zentner

et al., 2008). In this review, the term music-specific


refers to studies that employ just such a model of
emotions. As in the case of research approaches, there
is often overlap, and any study may utilize multiple
models for emotions. Thus, where necessary, a study
will be coded as featuring several emotion models. In
addition, there is an overlap between the classes of
emotion models, since the components of a musicspecific model may be either discrete or dimensional,
and basic emotion terms may be operationally defined
to form dimensions (e.g., happy-sad continuum). The
purpose of the model division is to emphasize the
epistemological sources of the emotion models.
LOCUS OF EMOTION

The locus of emotioncoined by Evans and Schubert


in 2008indicates whether the study addresses felt
emotions (also known as induced emotions, or internal
locus) or perceived emotions (external locus). This difference between the emotion actually experienced by
the listener (hereafter, felt) and the emotion recognized
in the music, (or expressed by music; hereafter, perceived) is of fundamental importance (Gabrielsson,
2001). A difference in the locus of emotion may sometimes lead to widely differing results (Konecni, Brown,
& Wanic, 2008; Zentner et al., 2008), even if the occasional, direct comparisons between felt and perceived
emotion have not revealed significant differences
between the two (see Evans & Schubert, 2008; Vieillard
et al., 2008). It is probable that largely different underlying processes are involved in the two kinds of emotional locus. Moreover, accurately capturing felt
emotions is known among the research community to
be a challenging task (see e.g., Roy, Mailhot, Gosselin,
Paquette & Peretz, 2008). The locus of emotion was
classified as internal if the instructions stated that the
participants should focus on their own emotional
experiences or on the emotions that the music induces
or arouses in them. This was in contrast to directions
that emphasized the perception and recognition of
emotions represented or expressed by music (external
locus).
MUSIC STIMULI

The music used as stimulus material in the studies was


annotated in terms of total number of pieces, duration,
as well as the musical genres they represented. We also
coded the nature of the stimuli and how the initial
selection of the stimuli was carried out. Regarding the
number of stimuli, it was the number of separate experimental stimuli rather than the number of individual
compositions/songs that was counted, as often only

This content downloaded from 129.240.0.83 on Mon, 6 May 2013 08:34:26 AM


All use subject to JSTOR Terms and Conditions

Review of Music and Emotion Studies

a few compositions (two or three) were manipulated in


terms of tempo or register to create dozens of variants
for the experiment (e.g., Husain, Thompson, & Schellenberg, 2002; Ilie & Thompson, 2006). Therefore, our
estimation of the stimulus quantity is a generous estimate, because the variants of the same musical source
are calculated as separate stimuli.
The musical genres featured in the studies were categorized into classical, mixed, pop/rock, ethnic, film
music, and custom-made. The classical genre contained
any period of classical music (baroque, classicism,
romantic, and modern). The mixed category was used
when the experiment featured several distinct genres
(e.g., classical, jazz, pop). Ethnic music was appropriate
for cross-cultural studies that have employed nonWestern musical excerpts. We also tabulated whether
the stimuli were natural (i.e., real recordings) or artificial excerpts (i.e., synthetically generated or isolated
aspects of music such as sound samples or MIDI renditions), and what musical aspects were actually manipulated. We also catalogued whether the stimuli were
chosen by the researchers, by a previous study, by
a panel of experts, or by the participants themselves.
Due to the high demands of controlling for possible
confounds, most music and emotion studies probably
rely on experimenter-chosen stimuli. However, the
method for selecting the stimulus material is probably
of great importance, especially for studies focussing on
felt emotions. It is known that participant-selected
music is more effective when music is used as a diversion
in painful operations (Pelletier, 2004; also MacDonald
et al., 2003), and more likely to give rise to chill experiences (Blood, Zatorre, Bermudez, & Evans, 1999;
Grewe, Nagel, Kopiez, & Altenmuller, 2007a) than
music chosen by researchers.
Finally, we made an effort to assess whether the participants were likely to have been exposed to the
excerpts beforehand. Certain canonical classical music
pieces (e.g., Albinonis Adagio) are typical examples of
excerpts that the participants might already be familiar
with. The coding of all the variables was done by the
authors in the following manner: after the initial review
phase, the final coding categories and terminologies
were decided. Then each author coded over half of the
whole corpus, and the discrepant items in the overlapping studies were discussed. After this, the coding
was slightly revised and both authors double checked
the part that had been coded by the other. The only
truly subjective measure was the issue of familiarity
with the stimuli, in which case we had to rely on our
own intuitions rather than on objective accounts of
exposure.

311

Results
EMOTION MODELS: AN OVERVIEW OF THE RESULTS

Discrete emotion models. The principal findings of


the studies regarding discrete emotions show that emotions such as happiness, anger, sadness, and tenderness
are accurately perceived in music, even by such special
populations as children (e.g., Adachi & Trehub, 1998;
Dalla Bella, Peretz, Rousseau, & Gosselin, 2001; Nawrot,
2003), members of different cultures (Balkwill &
Thompson, 1999; Fritz et al., 2009), and braindamaged patients (Gosselin et al., 2005; Peretz, Gosselin, & Bouchard, 1998). Most of these discrete emotions
may also be induced in listeners by using particular
music, which also result in discriminable patterns of
physiological (Baumgartner, Esslen, & Jancke, 2006;
Etzel, Johnsen, Dickerson, Tranel, & Adolphs, 2006)
and neural responses (Schmidt & Trainor, 2001). These
emotions are the most frequently featured (>30%) in
music and emotion studies, and the majority of these
studies utilized a self-report approach, the dominant
approach in this field. The popularity of discrete emotions is without doubt related to the fact that categories
lend themselves easily to recognition paradigms
common in the above mentioned developmental,
cross-cultural, and biological studiesbut are also frequently used in production-recognition studies (e.g.,
Gabrielsson & Juslin, 1996; Juslin, 1997a; Laukka &
Gabrielsson, 2000). Interestingly, investigations focussed
on exploring the relationships between emotions and
individual factors (Kreutz, Ott, Teichmann, Osawa, &
Vaitl, 2008; Resnicow, Salovey, & Repp, 2004), clinical
states (Banich, Stolar, Heller, & Goldman, 1992), or
visual cues in emotion recognition (Baumgartner et al.,
2006), adopt discrete emotions more frequently than
other emotion models. This is probably another aspect
of how concepts that are known to be easily discriminated and explained make their way easily into subsequent studies of emotions despite the epistemological
assumptions and limitations related to them.
The main difference between discrete emotion
research in general and in music and emotion studies
is that in the context of music the emotion categories
have typically been modified by replacing musically
inappropriate categories such as disgust and surprise
with more fitting categories such as tenderness or peacefulness (e.g., Balkwill & Thompson, 1999; Gabrielsson &
Juslin, 1996; Vieillard et al., 2008). This is perhaps indicative of the emotional functions of music that may generally be more positive (Juslin & Laukka, 2004; Zentner
et al., 2008) than the emotions experienced in everyday
life situations (i.e., basic emotions that emphasize the

This content downloaded from 129.240.0.83 on Mon, 6 May 2013 08:34:26 AM


All use subject to JSTOR Terms and Conditions

312

Tuomas Eerola & Jonna K. Vuoskoski

survival aspects of affects). Accordingly, the choice of


emotion categories may be too limited, and the impressive set of results could be interpreted as mainly showing
a broad affective distinction between positive and negative emotions, thus being of limited value to a deeper
understanding of emotions in music. Another significant
limitation is the focus on a mixed and narrow palette of
discrete emotions, which is not usually theoretically
driven. For instance, over 75% of the studies adopting
the discrete emotion model utilize happiness, sadness,
and anger. The remainder of the emotion categories
across studies often consist of different emotion concepts
such as peacefulness, fear, joyfulness, solemnity, tenderness, distress, disgust, and surprise. This seriously limits
the comparability of the results across studies.
Another sobering observation is that the results concerning several emotion categories could also be characterized by other emotion models to equal or better
degree. For example, the discrimination of happy and
sad emotions could also be delineated by differences in
valence and arousal. Also, reliance on forced-choice
paradigms and the presentation of a low number of
emotion categories may both lead to overestimating the
recognition success of these emotions. Finally, discrete
emotion studies have excessively concentrated on perceived emotions, and the relationships between these
and induced emotional experiences are often unclear
(Gabrielsson, 2002; Scherer & Zentner, 2001). Keeping
in mind these caveats and the ongoing debate about the
uniqueness of the physiological mechanisms concerning discrete emotions (e.g., Barrett & Wager, 2006), as
well as the disagreement about the exact number and
labelling of emotion categories, we should regard the
results on discrete emotions as the first exploratory
phase of emotion and music studies.
Dimensional emotion models. During the last two
decades, about a third of music and emotion studies
have utilized dimensional emotion models. Valence and
arousal are by far (> 70%) the two most typical dimensions that participants are asked to rate independently
on bipolar scales. The recognition of musical excerpts
that are either high or low on valence or arousalas
well as the four combinations of these bipolar
extremesseems to be effortless for listeners (Ilie &
Thompson, 2006; Leman, Vermeulen, De Voogdt, Moelants, & Lesaffre, 2005). Affects representing the same
four quadrants may also be induced by musical excerpts
(Rickard, 2004; Vieillard et al., 2008), and show unique
patterns of neural (Altenmuller, Schuermann, Lim, &
Parlitz, 2002; Kabuto, Kageyama, & Nitta, 1993) and
physiological responses (Gomez & Danuser, 2004;
Witvliet & Vrana, 2007)although these seem to be

more easily attributed to the arousal dimension than


to the valence dimension. In our view, this probably
relates to the special nature of musical experiences
where negative emotions are less common (cf. Juslin
& Laukka, 2004). Affects typically classified as negative
(e.g., melancholy) may be experienced as positive in
the context of musicdue to the fact that music does
not have any material, negative effects on the listeners
well-being.
In comparison to general studies of emotions, variant
formulations of the dimensional emotion model have
received little attention in music-related research,
despite the common observations that valence and
arousal are not able to account for all the variance in
music-mediated emotions (Bigand, Vieillard, Madurell,
Marozeau, & Dacquet, 2005; Collier, 2007; Ilie &
Thompson, 2006; Leman et al., 2005; Nyklcek, Thayer,
& Van Doornen, 1997). This lack of resolution is also
manifested in the way the two-dimensional model
places emotions commonly regarded as remote in close
proximity in the affective space; e.g., boredom and
melancholy both are negatively valenced and low on
arousal despite their notable differences in character
(Scherer, 2004). These observations have led emotion
researchers to seek other formulations of the dimensions (e.g., Schimmack & Grob, 2000; Thayer, 1989),
but surprisingly few of these developments have found
their way into music and emotion studies; only a handful of studies have utilized other dimensions such as
intensity, dominance, or tension in addition to the
valence and arousal dimensions (e.g., Ilie & Thompson,
2006; Leman et al., 2005; Luck et al., 2008; Schubert,
2007). The most critical flaw in past work concerning
the use of dimensional emotion models in music, however, is the seemingly tenacious confidence in Russells
circumplex model (valence and arousal), despite the
development of more empirically and neurologically
convincing dimensional models in the late 1980s. For
example, the Positive and Negative Affect Scale
(PANAS by Watson, Clark, & Tellegen, 1988) treats
positive and negative valence as separate dimensions,
and Thayer (1989) divides the affect space into dimensions of energy arousal and tension arousal. It is difficult
to avoid the impression that a few popular studies (Nyklcek et al., 1997; Schubert, 1999; Terwogt & Van Grinsven, 1991) established the research agenda for the
following decade, or that the field has suffered from
a lack of theoretical reflection in the choice of emotions
in general.
Miscellaneous emotion models. The prevalence of
miscellaneous emotion models (66/251) can be
regarded as a reaction of the research community to

This content downloaded from 129.240.0.83 on Mon, 6 May 2013 08:34:26 AM


All use subject to JSTOR Terms and Conditions

Review of Music and Emotion Studies

many of the criticisms leveled at the discrete and dimensional models, outlined above. The concepts belonging
to this category (intensity, preference, similarity, emotional expressiveness, and tension) seek to characterize
aspects that are ignored by the discrete and dimensional
models, although some of these concepts are present
(but may have a different definition) in the other
formulations of the dimensional model (e.g., tension).
There is no single emerging trend in the results
concerning miscellaneous emotion models. The most
frequent concept, preference/liking (14 out of 66 studies) is of borderline interest to emotions, although theoretically they are often considered as the simplest form
of affect manifestations (Zajonc, 1980; for a discussion
regarding preference in the context of music, see
Scherer & Zentner, 2001). In some studies liking/preference has been used as a substitute for valence (e.g.,
Brown, Martinez, & Parsons, 2004). This idea was
empirically tested by North and Hargreaves (1997).
Although valence and preference have been observed
to correlate highly, most studies acknowledge the difference between the two (Eerola & Vuoskoski, 2011;
Rawlings & Leow, 2008; Ritossa & Rickard, 2004). Similarly, studies claiming to investigate emotional intensity may actually be measuring something that
constitutes the arousal dimension of the circumplex
model (e.g., Schmidt & Trainor, 2001). Studies devoted
particularly to emotional intensity (Brittin & Duke,
1997; Krumhansl, 1998; Lehmann, 1997) or expressivity
(Kamenetsky, Hill, & Trehub, 1997; Sloboda & Lehmann, 2001) avoid the semantic labelling problem that
concerns discrete andto a certain degreedimensional models. The results of these studies have suggested that intensity/expressivity differences can be
both induced and perceived in a consistent manner.
However, both these concepts collapse all emotional
reactions into a single uniform dimension, and despite
being descriptive and operational for certain purposes,
they seem unlikely to contribute to a deeper understanding of emotions in music.
Finally, we consider the evaluation of tensioneither
in music or as an experienced emotionattractive due
to its links with mood (POMS by McNair, Lorr, &
Droppleman, 1981), physiological measurements (i.e.,
muscle tension), prevalence in everyday descriptions
of musical expressivity (Juslin & Laukka, 2004), and
apparent relevance to music theoretical discourse
(Meyer, 1956; Lerdahl, 2001). However, tension has also
been observed to correlate highly (negatively) with
valence (Eerola & Vuoskoski, 2011; Krumhansl, 1997).
Also, the drawback of most studies on tension has been
the lack of a common conceptual background.

313

Exceptions to this are the studies by Ilie & Thompson


(2006), implementing the three-dimensional emotion
model proposed by Schimmack and Grob (2000), and
Zentner and his colleagues (2008). The latter relates to
the realm of music-specific emotions, summarized next.
Music-specific models. Though only a minority of
studies (19) have employed music-specific emotion
models, the results are of special interest sinceas the
emotion models are especially constructed for musical
purposesthese studies avoid some of the criticisms
levelled at the emotion concepts mentioned above. Hevners (1936) ordering of emotions along the circumference of a circle was the first music-specific model that
was further elaborated by several authors in subsequent
years (e.g., Farnsworth, 1954; Wedin, 1972). Contemporary formulations of music-specific emotions have all
suggested substantially more factors to emotional
experiences than those offered by popular dimensional
models. Asmus (1985) provided 9 dimensions, while
Juslin offered over 15 emotions that exceed prevalence
of 70% (Juslin & Laukka, 2004; see also Juslin, Liljestrom, Vastfjall, Barradas, & Silva, 2008; Laukka, 2007),
and Zentner settled on nine emotion factors that may
also be collapsed into three metafactors (Zentner et al.,
2008). Encouragingly, these recent music-specific models share many factors (feeling moved, nostalgic,
relaxed, enchanted), and offer uniquely aesthetic
emotions.
The most important observation from these studies is
that the emotions induced by music are largely different
from those usually studied under the label of basic emotions or the common dimensional models. Even if some
of the emotions in the music-specific models are
labelled in the same way as in the common discrete
models, musical versions of these emotions seem to
have a different character. For example, sadness in
music is not usually associated with the aversive aspects
of the emotion such as feeling gloomy, depressed, or
unhappy (Laukka, 2007; Zentner et el., 2008). Note that
the situation seems to be more straightforward in
expressed emotions since no evidence has been provided that music-specific emotions would be easier to
perceive or express in musical performances than the
common discrete emotions.
RESEARCH APPROACHES: A SUMMARY OF THE RESULTS

Theoretical approaches. Theoretical approaches have


mainly acted as a catalyst, serving the field as a test bed of
ideas (e.g., Gabrielsson, 2002; Juslin et al., 2001; Konecni,
2008; Schubert, 1996; Sloboda, 2001), promulgating concepts from other fields (e.g., London, 2001; Panksepp,
1992; Scherer, 2004), or providing meta-analyses (Juslin

This content downloaded from 129.240.0.83 on Mon, 6 May 2013 08:34:26 AM


All use subject to JSTOR Terms and Conditions

314

Tuomas Eerola & Jonna K. Vuoskoski

& Laukka, 2003; Vastfjall, 2002). The latter studies aside,


perhaps the most influential theoretical studies have
focused on specific, theoretical problems such as the
dilemma of felt vs. perceived emotions (Gabrielsson,
2002), enjoyment of negative emotions (Schubert,
1996), and the proposal regarding a variety of mechanisms through which music may induce emotions (Juslin
& Vastfjall, 2008). These studies have all dealt with topics
that seem to operate differently from mainstream emotion psychology. Somewhat surprisingly, studies that
bring insights from non-music topics do not seem to
have made such an impact on this field (e.g., many of
Scherers studies). It needs to be said that regarding the
theoretical aspects of music and emotions, the handbook
of music and emotions by Juslin and Sloboda (2001) is
probably the most influential theoretical source and
certainly the most cited reference in papers published
after 2001.
Self-report approaches. The self-report approach has
provided the cornerstone of the results in music and
emotion research. It is the only way the listeners themselves can actively characterize their own, subjective
emotional experiences. The dominant self-report
measure has been the rating of an emotion construct
on a Likert scale, whichdepending on the construct
measuredis known to produce reliable estimates of
emotions across participants (e.g., Juslin, 1997a,
1997b). The results obtained in this manner are also
linked with the affects acquired with more openended response formats such as adjective choices and
free responses (Juslin, 1997a; Schubert, 2003).
The drawbacks of the self-report approach have rarely
been questioned and it is more customary to see selfreports as a baseline to which physiological, acoustical,
or neural measured are being compared. Most of the
self-report measures are reduced verbalizations of the
emotional experiencethough exceptions such as
similarity judgements existwhich are subject to limitations such as demand characteristics, and limitations
in ones awareness and ability to explain emotions. The
forced-choice paradigm is also a potentially valid selfreport method provided that the biases related to choosing between a finite set of categories (i.e., guessing,
representation of mixed emotions) are overcome.
Biological approaches. The biological approach has
concretely demonstrated that if the emotions are sufficiently different (i.e., sadness, happiness, anger, or distinct quadrants of the affect space), musically induced
emotions do lead to discriminable physiological and
neural patterns. At the physiological level, high arousal
is related to respiratory activity (Nyklcek et al., 1997)
and perspiratory activity (Rickard, 2004). Differences in

emotional valence may be differentiated by the activation of various facial muscles (Witvliet & Vrana, 2007),
including the startle reflex (Roy et al., 2008). However,
these observations are often conflicting and indeterminate (for example, Etzel et al., 2006; Nater, Abbruzzese,
Krebs, & Ehlert, 2006), if not downright trivial concerning the generic arousal dimension. When the focus of an
emotion study is on strong responses, it has become
customary to record chill reactions (Blood & Zatorre,
2001; Grewe et al., 2007a) in order to verify the temporal locations and the existence of such experiences in
participants.
Studies measuring neural responses have documented
the areas that are involved in emotional processing by
investigating musically induced emotions in healthy
participants (Blood & Zatorre, 2001; Koelsch et al.,
2006). Several studies have corroborated the discrimination of differentially valenced emotions by means of
neural responses, as different cortical lateralization patterns emerge in response to positive and negative musical stimuli (Altenmuller et al., 2002; Flores-Gutierrez
et al., 2007; Schmidt & Trainor, 2001). In our view, the
most intriguing contributions have come from studies
with brain-damaged patients, showing selective emotional impairments concerning with, for example, the
processing of fear (Gosselin, Peretz, Johnsen, &
Adolphs, 2007), implying specificities in the emotional
processing of music. The most influential1 neurological
study on music and emotions has undoubtedly been the
one that likened the emotional and rewarding effects of
preferred, chill-inducing music with strong, euphoriainducing stimuli such as food, sex, and drugs (Blood &
Zatorre, 2001), and it seems to have generated a series of
chill-inducing studies in its wake.
The use of biological measures has become frequent,
but the analysis methods and self-report methods have
not developed at the same pace. For instance, many of
the studies involving peripheral and neural measures
reduce the data into static segments compatible with
static self-report measures instead of capturing the
dynamic aspects of music-induced responses that would
require more sophisticated time-series analyses (see
Korhonen, Clausi, & Jernigan, 2006).
Developmental approaches. The developmental
approach has documented the ages at which children
are able to detect particular emotions in music, and the
emotional cues they are responsive to at each point of
1

Cited 256 times in ISI Citation Index, which is above the other
neuroscience papers concerning music and emotions, e.g., Peretz et al.
(1998): 86; Koelsch et al. (2006): 75; Brown et al. (2004): 48; Gosselin et al.
(2005): 36; Khalfa et al. (2002): 33; Altenmuller et al. (2002): 22.

This content downloaded from 129.240.0.83 on Mon, 6 May 2013 08:34:26 AM


All use subject to JSTOR Terms and Conditions

Review of Music and Emotion Studies

their development. This approach has also been important in highlighting the aspects of emotions that are the
least culture-dependent. Sensitivity to emotional communication is thought to be present in early infancy due
to the importance of caregiver emotion communication
via speech and singing (Trehub & Nakata, 2001), and
this claim is supported by some empirical findings
(Nawrot, 2003). Most developmental studies have
focussed on the preschool years, describing the ability
to distinguish an increasing range of basic emotions
using forced-choice paradigms with pictorial representations of emotion categories. The results suggest that,
around the age of four, children are able to recognize
happiness and sadness in music (Cunningham & Sterling,
1988), as well as anger and fear (Boone & Cunningham, 2001), although these latter emotions are still
frequently confused by older children and even by
adults (Dalla Bella et al., 2001; Robazza, Macaluso, &
DUrso, 1994; Terwogt & Van Grinsven, 1991).
Throughout their development, children begin to
apply more and more style-specific musical cues
such as mode (major-minor) divisionin determining
the appropriate affect (Dalla Bella et al., 2001; Kastner
& Crowder, 1990).
Cross-cultural and individual approaches. The crosscultural and individual approaches are both rare, but
the extant cross-cultural studies have demonstrated that
at least the broad, discrete categories of emotions
(happy, sad, and scary expressions in Fritz et al., 2009,
and joy, sadness, and anger in Balkwill & Thompson,
1999) are communicated across cultures, mirroring the
findings of cross-cultural speech expression recognition
studies (e.g., Scherer, Banse, & Wallbott, 2001). The
underlying reasons for the successful communication
have been expressed by means of a cue redundancy
model (Balkwill & Thompson, 1999), which postulates
that a set of psychophysical as well as culture-specific
cues account for emotions.
The cross-cultural approach is an extremely powerful
way of looking at the division between universal and
cultural associations regarding music and emotion.
Having said that, such studies are particularly challenging to carry out in a convincing way. Most of these
studies focus on emotion recognition, probably because
of the same challenges that are also present in developmental studies where forced-choice paradigms and
pictorial representations of emotions are utilized in
order to avoid verbal descriptions. In cross-cultural
studies, a more important concern is the conceptual
equivalence of the studied emotional concepts as well
as translation issues, which may both lead to unfounded
comparisons. Also, it is nearly impossible to find

315

participants in the contemporary world who would not


have any prior exposure to Western music. In this
sense, the study by Fritz and colleagues (2009) that
featured Mafa listeners from remote settlements in
the North Province of Cameroon can be considered
invaluable.
The studies utilizing an individual approach to emotions have sought to assess the role of musical expertise,
gender, or personality traits in emotion recognition or
experience. Most studies have failed to find major
differences in emotional processing stemming from
musical expertise or gender (Bigand et al., 2005; Frego,
1999; Juslin, 1997c; Kreutz, Bongard, & Von Jussis,
2002), although exceptions to this observation exist
(Fredrickson, 1995; Hoshino, 1996), particularly concerning psychophysical reactivity in men and women
(Nater et al., 2006). The stable personality traits of
listeners have been observed to influence emotional
reactions in ways predicted by such theoretical assumptions as trait-congruency (i.e., individuals preferentially
process stimuli that are emotionally congruent with
their personality traits; see e.g., Kallinen & Ravaja,
2004; Vuoskoski & Eerola, 2011). To date, studies investigating individual factors have been limited to prevalent emotion models (basic emotions, or valence and
arousal), and the role of individual factors has been
moderate in comparison with the overall emotion recognition accuracies. In future studies, it would be interesting to investigate the extent to which overall
emotional intensity, tendency to represent emotions
with precision and specificity (Tugade, Fredrickson, &
Barrett, 2004), and everyday uses of music are linked
with individual factors.
Music analytic approaches. The music analytic
approach has documented the wide range of reliable
cues that characterize emotions in music. Production
studieswhich can be divided into studies of emotional
expressivity (Juslin, 1997c; Laukka & Gabrielsson, 2000)
and composition (e.g., Thompson & Robitaille, 1992;
Vieillard et al., 2008)have produced the largest impact
on our understanding of the features related to emotions. A stream of analytical studies, where emotioninducing musicverified by self-reportsis dissected
into its constituents by means of acoustic and musical
analyses or rating studies (e.g., Leman et al., 2005; Schubert, 2004) have reinforced the results of the production
studies. From all of these studies, systematic couplings
between musical features and particular emotions have
been established. For instance, happy expression is
achieved by fast voice onsets, bright timbre, major
mode, and fast tempo, whereas anger has otherwise
similar features but loudness tends to be higher and the

This content downloaded from 129.240.0.83 on Mon, 6 May 2013 08:34:26 AM


All use subject to JSTOR Terms and Conditions

316

Tuomas Eerola & Jonna K. Vuoskoski

FIGURE 1. A summary of the approaches and emotion models (discrete/dimensional/ miscellaneous/music-specific) in reviewed studies. The
connectors display the prevalence of concurrent approaches, where the bold lines denote > 20 studies, solid line > 10 studies, and dashed line > 5
studies.

microstructure more irregular (cf. Gabrielsson & Lindstrom, 2010). Direct manipulations of the central features such as tempo, mode, and register have
corroborated their role in music and emotions (Ilie &
Thompson, 2006; Juslin, 1997c).
Despite the extensive, detailed reports of the featureemotion couplings, the scope of the studies has largely
remained neutral in terms of musical style (no particular style or resemblance to film soundtracks or classical
music). A great deal of further research is needed in
order to establish underlying similarities of these couplings across a larger variety of musical styles. Also, the
music analytic approach has been too rarely used in
conjunction with other approaches, despite the direct
gains such an approach could bring to the generalizability of the findings.

PREVALENCE OF RESEARCH APPROACH, EMOTION MODEL, AND


LOCUS OF EMOTION IN ALL STUDIES

To highlight the broad trends in an effective manner,


compact summaries depicting every research approach,
locus of emotion, and emotion model are first presented
in graphical form. The references for the studies that
belong to each category are given in the appendix
(Tables A1-A7 in the Appendix).
Figure 1 displays a simplified quantitative summary
of studies falling into the chosen seven research
approaches as well as the prevalence of the simultaneous usage of several approaches in a single study. For
instance, more than ten self-report studies have
collected data relevant for the individual and cultural
differences approach category. The numbers in brackets
refer to the number of each kind of emotion model used

This content downloaded from 129.240.0.83 on Mon, 6 May 2013 08:34:26 AM


All use subject to JSTOR Terms and Conditions

Review of Music and Emotion Studies

317

TABLE 1. Cross-Tabulation of Emotion Models (Rows), Approaches (Columns), and Locus of Emotion (in Subscripts Felt/Perceived/Both).

Discrete
Dimensional
Misc.
Music-specific
%*

Self-R.

Biol.

Mus A.

Theor.

Clin.

Ind./Cul.

Dev.

%*

8016/65/3
8140/45/4
5124/17/12
186/11/4
52%

238/10/5
3926/13/1
128/2/2
11/1/
17%

301/26/3
164/13/
159/3/4
2/2/
14%

52/3/1
63/3/1
74/1/2
21/1/1
5%

101/9/1
81/7/1
2/2/
1///1
5%

72/5/
65/1/
53/1/1
0//
4%

14/14/
1/1/
0//
1//1/
3%

38%
35%
21%
6%

*Proportion of total studies refers to number of studies directly classified according to research approach, and may differ from the row/column totals as all the studies do not
report the exact model utilized, and the several approaches and models may be used simultaneously.

within each research approach in the following order:


discrete/dimensional/miscellaneous/music-specific.
What is interesting in the visualized cross-tabulation
of the research approaches and emotion models in
Figure 1, is that while the self-report approach has been
the most common (77% of the studies), two other
approaches are equally common (biological and music
analytic, each about one quarter of the studies). The
self-report approach is probably slightly different in
nature compared to the other approaches, as it is the
hub upon which all other approaches rely in order to
validate that participants were actually experiencing the
intended emotions (when the measurements are biological). It must also be said that the link between theoretical studies and other approaches should be regarded in
a special way, since all empirical studies have made use
of theoretical approaches in their study designs and
hypotheses. Hence, the three studies showing explicit,
concurrent use of theoretical and self-report approaches
may be considered as exceptions; two-part articles with
the main emphasis on the theoretical section and
a proof of concept given in a shorter empirical section
(e.g., Holbrook & Anand, 1992; Juslin & Laukka, 2004;
Madsen, 1997a).
It is worth noting that special populations (featured in
clinical, developmental, and individual and cultural differences approaches) are rarely combined with anything
other than self-report approaches, probably due to their
reliance on special populations. Also, biological
approaches combined with individual and cultural
differences are relatively scarce because of the specificity
of the questions involved as well as for practical reasons
(for example, a field study in an isolated geographical
location with a non-Western influenced population
would not have easy access to a MEG recording device).
Figure 1 gives a fairly uniform message that the discrete and dimensional models remain the two most
prevalent emotion models in music and emotion
research. In the studies using self-report and clinical
approaches, these two models are equally common, but

in developmental and biological studies the balance is


somewhat different. In developmental studies, children
are typically asked to choose a suitable facial expression
from the two or three candidates offered to match the
emotion expressed by music, and hence the emotion
model is typically discrete (happy/sad). In biological
studies the underlying physiological mechanisms
involved are often characterized using valence and
arousal (with physiological or neurological explanations
pointing towards avoidance and approach systems),
thus directing the choice of emotion model towards the
dimensional. The miscellaneous and music-specific
emotion models are almost equally dispersed among
the variety of research approaches, suggesting that these
models can easily accommodate different methodological and conceptual approaches.
The prevalence of emotion models across all studies
and how often they are used simultaneously in a single
study is shown in Table 1. The two dominant (70%)
emotion models, discrete and dimensional, are used
almost equally frequently, and both are used in the same
study in about 10% of the studies utilizing these models.
This seems to be a recent trend, as most of the studies
combining discrete and dimensional models are published after the year 2004 (see Figure 2 for the details of
the temporal trends). As such, the use of two different
emotion models in the same study might reflect the
need to compare the models, or it may be seen as a sign
of acceptance that emotions may easily be conceived as
both belonging to discrete categories while also representing a dimensional continuum. Also, the number of
studies (14) not specifying the locus of emotion mostly
reflect the era before the issue was made explicit by
Gabrielsson (2002) and Sloboda (2001), since a large
proportion (57%) of these studies predate these two
landmark studies. Although the combined use of
music-specific and traditional emotion models in the
same study is not common, the miscellaneous models
co-occur with the dimensional models in 15 studies.
These studies often collect valence or arousal ratings,

This content downloaded from 129.240.0.83 on Mon, 6 May 2013 08:34:26 AM


All use subject to JSTOR Terms and Conditions

318

Tuomas Eerola & Jonna K. Vuoskoski

FIGURE 2. Temporal trends of the locus of emotion (upper panel) and emotion models (lower panel). The percentage for each category is also
displayed to allow better differentiation from the yearly increase of the pertinent studies.

liking, or preference ratings (Brown et al., 2004; Rawlings & Leow, 2008), and mood (Husain et al., 2002) or
intensity ratings (Kreutz et al., 2008).
Miscellaneous models are the third most common of
the four categories, and within these models, emotional
intensity, preference, and tension are the most common
choices for alternative self-report concepts. Musicspecific models are only featured in 19 of all 251 studies,
but if we look at these studies in detail, we notice that
only a few influential music-specific emotion models
have been proposed over the years. The first one is
Wedins (1972) model, which consists of three bipolar
factors (gaiety gloom, tension relaxation, solemnity
triviality). This model was used by only one of the
studies we looked at (Giomo, 1993). The second formulation of a music-specific emotion model was made by
Asmus (1985), who settled for nine affective dimensions
that have been used by at least three studies over the
past 20 years (Coffman, Gfeller, Eckert, 1995; Gfeller &
Coffman, 1991; Gfeller, Asmus, & Eckert, 1991). The

third music-specific emotion model was offered by Bartel (1992) and contained 12 bipolar dimensions. Recent
surveys of emotion terms associated with music (Juslin
& Laukka, 2004; Juslin et al., 2008; Lindstrom, Juslin,
Bresin, & Williamon, 2003; Scherer, Zentner, &
Schacht, 2002; Zentner et al., 2008) have greatly
enlarged the scope of emotion concepts available for
contemporary researchers. From this literature, the
Geneva Emotional Music Scale (GEMS) by Zentner and
his colleagues (2008) may be the most easily approachable and the most systematically constructed.
In terms of the locus of emotion, dimensional, miscellaneous, and music-specific models are evenly distributed
between emotions that are felt and perceived in music.
For discrete emotion models, there seems to be an
emphasis towards a perceived locus of emotion, probably due to the emotion recognition tasks that are popular in all developmental, cross-cultural, and some
biological studies (Table 1). It is also worth noting that
there has been a shift in focus during the last 6-8 years

This content downloaded from 129.240.0.83 on Mon, 6 May 2013 08:34:26 AM


All use subject to JSTOR Terms and Conditions

Review of Music and Emotion Studies

319

TABLE 2. Summary of Research Approaches and Emotion Loci in Journal Articles Addressing Music and Emotions During the Last Two Decades.

Felt Emotions

Perceived Emotions

Methodological Approach
Self-report
Juslin & Laukka (2004); Konecni et al. (2008)
Juslin (1997c); Ritossa & Rickard (2004)
Biological
Blood & Zatorre (2001); Witvliet & Vrana (2007) Khalfa, Roy, et al. (2008); Peretz, Blood, Penhune,
& Zatorre (2001)
Music analytic
Sloboda (1991)
Schubert (2004)
Theoretical
Scherer (2004); Sloboda (2001)
Scherer (1995)
Clinical
Khalfa, Delbe, et al. (2008)
Dellacherie et al. (2008); Nielzen, Olsson, &
hman (1993)
O
Individual differences Kallinen & Ravaja (2004); Nater et al. (2006)
Rawlings & Leow (2008)
Developmental
Dalla Bella et al. (2001); Terwogt & Van Grinsven
(1991)
Emotion Model
Discrete
Baumgartner et al. (2006); Krumhansl (1997)
Juslin (1997a); Kallinen (2005)
Dimensional
Grewe et al. (2007a); McFarland & Kennison
Gagnon & Peretz (2000); Ilie & Thompson (2006)
(1989a),
Miscellaneous
Bigand et al. (2005); Thaut & Davis (1993)
Brittin & Duke (1997); Frego (1999)
Music-specific
Dibben (2004); Zentner et al. (2008)
Bartel (1992); Gfeller et al. (1991)

from emotion perception towards music-induced, experienced emotions. This historical continuum has been
visualized in Figure 2, which displays temporal histograms of the publication years of the studies divided
according to the locus of emotion (upper panel) and
emotion model (lower panel). The locus of emotion
panel displays a slight positive trend towards the simultaneous investigation of felt and perceived emotions over
the last five years. As previously mentioned, this applies
equally to all emotion models: Figure 2 shows that using
several emotion models in one study has also become
more common recently. And finally, the prevalence of
both miscellaneous and music-specific models of emotion has increased during the last 6 years, although no
sudden emergences of a particular model are apparent.
SUMMARY OF POPULAR THEMES

To bring together the results regarding the prevalence of


each research approach, emotion model, and locus of
emotion, a cross-tabulated listing of these components
is shown in Table 1. The first number in each cell indicates the number of studies corresponding to each
research approach (column) as combined with each
emotion model (row). The small numbers separated
by forward slashes stand for the locus of emotion. They
indicate the number of studies falling into one of three
categories of locus of emotion (felt, perceived, both).
The full breakdown of the studies, indicated in Table
1, is given in the appendices (A1-A7).

The summary underlines the dominance of selfreport approaches ( 80% of studies) that use discrete
(43%) and dimensional (43%) emotion models. In general, the locus of emotion is evenly divided between felt
and perceived, although the prevalence varies between
research approaches and emotion models. Variations in
the balance between felt and perceived emotions are to be
expected, as the studies that emphasize biological
approaches typically focus on felt emotions, anddue
to limited instructional possibilitiesdevelopmental
studies have to rely on emotions that are perceived or
recognized by children. Also, within the discrete model
for emotions, perceived emotions have been favored due
to an emphasis on the recognition of emotion categories.
To give the reader a few textbook examples of studies that
reflect each research approach, emotion model, and locus
of emotion, two example studies for each approach (7)
and emotion model (4) are given for each locus of emotion
(2) in Table 2. The example studies have been selected to
represent a variety of years, and provide recommended
reading for those interested in such approaches.
MUSICAL STIMULI

Table 3 encapsulates selected details about the stimuli


used in music and emotion studies. Note that many
studies did not use any stimuli (or the reported details
were not accurate enough), and hence the table only
shows details from 170 of the 251 studies. Regarding the
question of musical genre, it was no surprise to discover

This content downloaded from 129.240.0.83 on Mon, 6 May 2013 08:34:26 AM


All use subject to JSTOR Terms and Conditions

320

Tuomas Eerola & Jonna K. Vuoskoski

TABLE 3. Summary of Stimulus Details From a Subset of Studies


(N 170).

Frequency (%)
Genre
Classical
Mixed
Pop/Rock
Ethnic
Film music
Custom-made
N/A
Nature of stimuli
Commercial recordings
Custom-made
Chosen by
Researchers
Pilot study
Previous study
Panel/Expert group
Participants
N/A

82 (48%)
27 (16%)
5 (3%)
6 (4%)
3 (2%)
18 (11%)
29 (17%)
115 (76%)
37 (24%)
57 (33%)
14 (8%)
16 (9%)
10 (6%)
7 (4%)
43 (39%)

that classical music is still the dominant genre (prevalence 48%), as confirmed by a previous meta-analysis
(41% in Juslin & Laukka, 2003). This was also observed
to be the prevalent musical style in a meta-analysis of
musical mood-induction studies by Vastfjall (2002, p.
176). Almost a fifth of the studies use several genres
(often three separate styles such as classical, pop, and film
music). In these cases, the stimulus selection is characterized either by such broad terms as drawn from the
corpus of Western music (Gomez & Danuser, 2004), or
by being selected from different musical styles to elicit
different types of emotional responses (Ali & Peynircioglu, 2006; Nater et al., 2006), or deliberately to make the
results as general as possible (Hunter, Schellenberg, &
Schimmack, 2008). Despite the increased number of
studies using mixed genres, it seems that further work
is needed in this area, as emotional reactions may differ
across genres due to their inherent musical differences
(i.e., the functions of music; e.g., Zentner et al., 2008). It
has also been discovered that music preferences influence
felt and perceived emotions (Kreutz et al., 2008). It is
somewhat surprising that many popular genres such as
R&B/soul, dance, or chart pop are featured rarely or not
at all in the studies on music and emotion (see Eijck,
2001; Mulder, Ter Bogt, Raaijmakers, Gabhainn, &
Sikkema, 2009; North, Hargreaves, & Hargreaves, 2004).

NATURE OF THE STIMULI: COMMERCIAL RECORDINGS VERSUS


CUSTOM-MADE EXAMPLES

The majority of the studies reviewed (75%) used


excerpts of real, commercially recorded music, which
at least dispels some of the notions of musical examples
being artificial and contrived. It should be kept in mind
that real and synthetic stimuli are typically used for
entirely different purposes. While the use of real excerpts
keeps the ecological validity high, such examples lack the
control of musical and acoustical parameters. In this
respect, artificially manipulated stimuli allow the
researcher to probe into predefined musical factors (such
as register, mode, tempo, and volume) in a controlled
manner. To make matters more complicated, precise
control of musical features is not always guaranteed
even when synthetic stimuli are employedif the manipulation scheme does not follow strictly defined factorial
parameterization of the chosen features.
Custom-made or synthetic stimulus materials are
often composed to sound natural to the listeners (e.g.,
the Montreal Battery, Peretz et al., 1998, addressed in
detail later), and such a creation principle will probably
not include systematic, factorial manipulation of musical parameters that would have enabled the researcher
to study the role of each manipulated feature separately.
In fact, systematic manipulation is usually not a common tactic in music and emotion research. The structural parameters of music have been manipulated in
terms of their symbolic representations by only a few
researchers (e.g., Dalla Bella et al., 2001; Gagnon &
Peretz, 2003), and likewise only a few studies have
manipulated performance features (e.g., tempo, register,
loudness) in terms of the audio representation of real
music recordings (Ilie & Thompson, 2006; also indirectly by Hunter et al., 2008). Hybrid forms of stimuli,
where the actual performances of simplified musical
materials (typically melodies with varying emotional
expression) are recorded, exist, and are used in emotion
recognition tests. In most production studies, high-level
professional musicians have been asked to perform simple melodies or rhythmic patterns in different ways to
express happiness, sadness, and tenderness (Gabrielsson & Juslin, 1996; Gabrielsson & Lindstrom, 1995;
Juslin, 1997c, 2000; Laukka & Gabrielsson, 2000; Timmers & Ashley, 2007). Variants of the task have invited
children to sing tunes with different emotional expressions (Adachi & Trehub, 1998, 2002), or composers to
create suitable music to befit the emotional expression
supplied by the researcher (Thompson & Robitaille,
1992). The production tasks in such custom-made samples provide a compromise between high control and
musical naturalness, and certainly this approach could

This content downloaded from 129.240.0.83 on Mon, 6 May 2013 08:34:26 AM


All use subject to JSTOR Terms and Conditions

Review of Music and Emotion Studies

be more commonly used in music and emotion studies


(cf. emotions expressed in speech, Banse & Scherer,
1996; Cowie & Cornelius, 2003).

321

possibilities to capture everyday emotional experiences,


and to relate them to the actual features of the music.
STIMULUS QUANTITY

STIMULUS SELECTION

The typical method of selecting the stimuli within the


field of music and emotion research has almost entirely
been researcher-driven: either the authors have intuitively selected the stimuli (33%), a pilot study has been
used, a panel or group of experts has selected the stimuli, or they have been chosen on the basis of a previous
study (8%, 6%, and 9%, respectively). In a minority of
the studies (4%), participants have been allowed to
choose the musical material themselves. This naturally
reflects the requirement to control the stimulus material, and the specific aims of individual studies to focus
on certain emotion models and approaches. To put it in
a simpler way: it makes sense to select arousing and
non-arousing music for a study that aims to determine
the extent to which emotional arousal induced by music
could be detected from peripheral measures (e.g., Nater
et al., 2006; Nyklcek et al., 1997), rather than to ask
participants to bring their own music. However, an
interesting but rare addition to such a tactic is to use
participant-chosen music in addition to researcherchosen music (Rickard, 2004), or to use only strongly
emotional music chosen by the participants themselves
(Blood & Zatorre, 2001).
The method of stimulus selection has undoubtedly
the largest impact on studies investigating felt emotions.
Music-induced emotions can be regarded as personal,
and they depend more on individual issues such as
learning, associations, and musical taste than the recognition of emotions in music (Juslin & Vastfjall, 2008).
As mentioned previously, in medically oriented studies
of pain and stress reduction (Pelletier, 2004), it has been
demonstrated that self-chosen music in general is more
efficient in modulating ones own mood than
researcher-chosen music (also in MacDonald et al.,
2003). Similar observations have been reported in studies that have aimed to investigate and record strong
emotional experiences induced by music (e.g., Blood
& Zatorre, 2001; Rickard, 2004). Thiscoupled
together with the notion that contemporary audio analysis and manipulation tools such as the MIR toolbox
(Lartillot & Toiviainen, 2007) or the Sonic Visualizer
(Cannam, Landone, Sandler, & Bello, 2006) greatly
facilitate controlling and describing participant-chosen
musicmeans that the use of participant-chosen music
is certainly a promising avenue for progress. Also, the
widespread popularity of portable music devices and the
way people use playlists will afford researchers new

We should now turn to the issues of stimulus quantity


and duration, which are presented as histograms in
Figure 3. The median number of stimuli in all studies
is 10, although the number varies substantially across
studies (from 1 to 100, see the upper panel in Figure 3).
While a large portion (66%) of the studies employs less
than 20 stimuli, there are also notable exceptions that
use 60 or more stimuli. Probably the stimulus number
itself is not always meaningful, as the design of the study
and subsequently the overall length of the experimental
session (< 60 min) are the overarching constraints on
the maximum number of stimuli used.
The typical stimulus duration, shown in the lower
panel of Figure 3, shows a pattern similar to typical
stimulus quantity (the upper panel): Short excerpts
(with a median of 60 s, 35% of the studies used excerpts
shorter than 30 s) are widespread, although 1-, 2-, and
3-min examples are also used in about one third of the
studies. As explained previously, stimulus duration
depends heavily upon the predetermined study design,
approach, and locus of emotion. For example, the
median stimulus duration of the studies focusing on felt
emotions is 90 s (with a mean of 135.4), while for studies focusing on perceived emotions this is 24 s (with
a mean of 62.8). The issue here is whether or not the
focus is on mere emotion recognition (perceived emotions), as this has been shown to be possible even when
extremely short musical clips (1 s in Bigand et al., 2005;
3.5-5 s in Altman, Alyanchikova, Guzikov, & Zakharova,
2000; 0.6 s in Goydke, Altenmuller, Moller, & Munte,
2004) are used. When the focus is on emotion induction
(felt emotions), the induction of an emotional response
and the subsequent self-reporting of the experienced
emotion may require more time. Judging from the typical stimulus durations used in studies investigating felt
emotions, it seems that 30 to 60 s is enough. Except for
the study by Altenmuller et al. (2002), which used 15-s
excerpts, very few similarly focused studies have
attempted to use excerpts shorter than 20-s. In direct
contrast, the decision to choose reasonably short
excerpts in most studies on perceived emotions reflects
the fact that the most common research approach used
in these studies is self-report (e.g., Likert ratings of the
emotion concepts in question), which typically require
the participant to give a single evaluation per excerpt.
For this reason, the musical examples are chosen to be
representative examples of a given emotion and do not
contain large temporal variations. Some of the longest

This content downloaded from 129.240.0.83 on Mon, 6 May 2013 08:34:26 AM


All use subject to JSTOR Terms and Conditions

322

Tuomas Eerola & Jonna K. Vuoskoski

FIGURE 3. Stimulus quantities and durations in a subset of music and emotion studies (N 170).

excerpts stem from studies that focus on the temporal


aspects of emotions using a continuous rating method
(McAdams et al., 2004; Schubert, 1999, 2003; Sloboda &
Lehmann, 2001).
To set the stimulus quantities of music and emotion
studies in perspective, a comparison with affective stimulus sets in other fields, such as the visual domain, is
worthwhile. The International Affective Picture System
(IAPS) uses 12 series of 60 pictures each (Lang, Bradley,
& Cuthbert, 1999), out of which typical emotion studies
tend to use 50-60 images (a median in a sample of 20
studies using IAPS). In comparison, the amount of
stimuli used in music-related studies is considerable
lower than this (median of 10, 66% used less than 20
stimuli). For emotions in speech, the stimulus sets tend
to have a larger number of stimuli, and the excerpts are
typically shorter (e.g., Burkhardt, Paeschke, Rolfes,
Sendlmeier, & Weiss, 2005; Douglas-Cowie, Campbell,
Cowie, & Roach, 2003).

emotion induction mechanisms such as evaluative conditioning and episodic memories (Juslin & Vastfjall,
2008). Familiarity also increases liking (Schellenberg,
Peretz, & Vieillard, 2008), and while that is a positive
factor, familiarity may confound the results of a study if
all participants are not equally familiar with the musical
examples used. Thus, we assessed whether the participants of the reviewed studies were likely to have been
exposed to the musical excerpts used. Although the
issue of probable prior exposure cannot be objectively
determined afterwards, we will offer examples of wellknown pieces of music that we think typical, educated
participants might be familiar with.
We were able to find 43 studies in which rather wellknown musical examples from the popular repertoire
had been used. For example, Tomaso Albinonis Adagio
in G minor for Strings and Orchestra had been used in at
least 16 studies.2 Other popular choices have included
Samuel Barbers Adagio for Strings, Gustav Holsts Mars
The Bringer of War, Modest Mussorgskys Night on

STIMULUS FAMILIARITY

Prior exposure to (or familiarity with) the music used as


stimulus material is a problematic factor for music and
emotion studies, as associations to television commercials, films, or past events can shape the emotional
experiences in unwanted or at least uncontrolled ways
(Juslin & Vastfjall, 2008). This issue is related to certain

Dalla Bella et al. (2001); Gosselin et al. (2006, 2007); Khalfa, Delbe,
et al. (2008); Konecni et al. (2008); Kreutz et al. (2008); Krumhansl
(1997); Lesiuk (2005); Martin & Metha (1997); Mitterschiffthaler, Fu,
Dalton, Andrew, & Williams (2007); Peretz et al. (2001); Peretz et al.
(1998); Schellenberg, Nakata, Hunter & Tamoto (2007); Tan,
Spackman & Bezdek (2007); Thompson, Schellengberg, & Husain (2001).

This content downloaded from 129.240.0.83 on Mon, 6 May 2013 08:34:26 AM


All use subject to JSTOR Terms and Conditions

Review of Music and Emotion Studies

the Bare Mountain, Antonio Vivaldis La Primavera


from The Four Seasons, Wolfgang Amadeus Mozarts
Eine Kleine Nacht Musik, Carnival of Animals by
Camille Saint-Saens, parts from the Peer Gynt Suite
by Edward Grieg, Giuseppi Verdis Brindisi from La
Traviata, Beethovens Symphony No. 6, Blue Danube
by Richard Strauss, Chanson du Toreador from Carmen
by Bizet, Wolfgang Amadeus Mozarts Requiem
(selected parts), and so on. A related issue is that many
studies do not explicitly report the exact movements or
segments used as stimuli, thus making their findings
non-replicable.
Many of these aforementioned, well-known pieces are
first used in one influential study (e.g., Krumhansl,
1997; Peretz et al., 1998) that gets cited in subsequent
studies that are either inspired by the initial study (e.g.,
Gosselin et al., 2006; Khalfa, Roy, Rainville, Dalla Bella,
& Peretz, 2008), attempt to replicate some of the findings (e.g., Konecni et al., 2008), or want to rely on the
safe musical choices established by a previous study
(e.g., Baumgartner et al., 2006; Schubert, 2004; Zentner
et al., 2008). The number of studies (42) featuring this
popular repertoire is considerably high considering that
there were only 81 studies with classical music recordings overall in the 251 reviewed studies. In the future,
studies investigating music-induced emotion should
avoid such overfamiliar stimulus materials, and hence
reduce possible unwanted associations.
As pointed out in the above summary of the popular
repertoire, there are relatively few originators of stimulus sets. Isabelle Peretz and her team in Montreal have
been extremely active in emotion studies in general, and
have launched two influential stimulus sets for music
and emotion research. The first contains 32 classical
music excerpts (Peretz et al., 1998) consisting of recordings of the popular classical repertoire (both orchestral
and solo piano works), most of which have worked their
way into a host of subsequent studies.3 The second set,
called the Happy, sad, scary and peaceful musical
excerpts for research on emotions (Vieillard et al.,
2008; documented also in Gosselin et al., 2005), consists
of 56 specifically composed, MIDI-generated synthetic
examples divided into four discrete emotion categories
(happy, sad, scary, and peaceful), which have been used
in at least nine studies to date.4 In mood induction
studies, some early authors (e.g., Albersnagel, 1988;
3
Dalla Bella et al. (2001); Fritz et al. (2009); Gosselin et al. (2007);
Khalfa et al. (2002); Peretz & Gagnon (1999); Schellenberg et al. (2008).
4
Dalla Bella et al. (2001); Gosselin et al. (2005, 2006, 2007); Khalfa
et al. (2002); Khalfa, Delbe, et al. (2008), Peretz & Gagnon (1999); Peretz
et al. (2001); Schellenberg et al. (2008); Vieillard et al. (2008).

323

Clark & Teasdale, 1985) have also created stimulus sets


that have been used by a large number of subsequent
studies (see Vastfjall, 2002, for details).
CONTEXTUAL AND METHODOLOGICAL ISSUES

Although we did not develop a coding for the contextual


aspects of music and emotion studies, some of these
may be briefly outlined in order to complete the picture
of the state of research. Almost all of the 251 studies
reviewed have been conducted in laboratory settings. In
a typical study, participants individually listen to and
rate the musical examples using headphones and a computer interface. Such situations are probably not the
most conducive to emotion induction, but such a controlled environment also ensures effective comparability
between listeners. We do not see the laboratory experiments as problematic for perceived emotion studies, as
emotion recognition is probably little affected by the
artificiality of the music listening situation. Only a few
studies have used more realistic contexts (mainly survey
studies; Juslin & Laukka, 2004; Zentner et al., 2008), but
the bolder attempts to measure emotions outside the
laboratory are worth mentioning, as they have led to
a new form of the self-report approach. For example,
Vaitl, Vehrs, and Sternagel (1993) had listeners rate
their emotional arousal while listening to selected leitmotifs from Wagners operas during the Bayreuth
Opera Festival. This idea was extended by McAdams
et al. (2004) by inviting concertgoers to continuously
rate their felt emotional force at a concert premiere.
A dominant portion of the studies using a self-report
approach rely on retrospective ratings of emotions, in
which a rating/ratings are provided after a stimulus has
been heard. However, music unfolds in time, and emotional experiences also change as a function of time. To
capture the moment-by-moment fluctuations in emotional experiences, continuous self-report measures
have occasionally been used (for example, Grewe
et al., 2007a; Luck et al., 2008; Schubert 1999, 2001,
2004; Sloboda & Lehmann, 2001). The disadvantages
of using continuous measurements are the extra effort
it takes for the participants to concentrate on the temporal fluctuations in their emotional experiences, and
the more complicated analysis techniques needed to
deal with such data. Physiological and neural measures
arealmost by definitiontemporal in nature, and in
these cases the temporal dynamics involved are usually
better developed. However, direct comparisons
between continuous self-reports and physiological
measures have not been common (e.g., Grewe et al.,
2007a; Krumhansl, 1997). In our opinion, nonverbal,
behavioral measures are too infrequently used to

This content downloaded from 129.240.0.83 on Mon, 6 May 2013 08:34:26 AM


All use subject to JSTOR Terms and Conditions

324

Tuomas Eerola & Jonna K. Vuoskoski

measure music-induced emotions. If we exclude the


forced-choice response formats such as facial pictures
representing different basic emotions, only isolated
studies have applied nonverbal indicators of emotions.
Two studies (Bigand et al., 2005, and Vieillard et al.,
2008) have collected similarity ratings for emotional
music, and thereby gained important insights into the
ways in which people organize emotional music examples. As such, this method has the same discovery
value as diary studies, which allow one to uncover such
patterns between music, emotions, and everyday life
that may not be detectable using theory-driven observations (such as self-reports with closed-response formats). Although some studies have tried to analyze
movements during music-induced emotions (Boone &
Cunningham, 2001; Camurri, Lagerlof, & Volpe, 2003),
or let the participants choose appropriate colors to
match emotions (Bresin, 2005), cross-modal issues of
music-induced emotion are still largely unexplored.
In most studies, lyrics are avoided in order to eliminate specific meanings generated by the verbal content.
For example, instrumental classical music has been the
musical style most commonly used, while jazz and film
music also do not usually have lyrics. Only a few studies
have looked into the question of how lyrics affect
emotional meaning in music (Gfeller & Coffman,
1991; Stratton & Zalanowski, 1994; Sousou, 1997). For
example, the findings of Ali and Peynircioglu (2006)
suggested that the musical content (melody in this case)
has more power to moderate emotions than lyrics,
although lyrics do seem to enhance negative emotions
more easily than positive emotions. This issue could and
should be addressed further, as in the context of everyday music consumption the question of how emotions
are generated by lyrics is of paramount importance.
Recommendations for Future Research

A number of problematic notions from past music and


emotion studies were brought up by the review. Several
of these were already identified by Juslin and Laukka in
their meta-analysis of discrete emotion studies in 2003,
and many of these challenges are also the reason for the
verdict on the confusing and inconsistent state of affairs
given by Juslin and Vastfjall in 2008. Here, however, we
outline the major challenges based on a wider variety of
music and emotion studies (251), and suggest concrete
plans of action for improvement.
THEORETICAL COMMITMENT

Perhaps the most critical shortcoming in the field has


been the reluctance to commit to the underlying

theories of emotions. Many of the concepts utilized in


music and emotion studies (e.g., fear, empathy, arousal)
carry much stronger implications of the underlying processes than are acknowledged or advocated in empirical
studies. The theoretical discussions are rare, and the
rare exceptions (e.g., Juslin & Vastfjall, 2008) have
therefore a large impact on the community. In practical
terms, the most significant source of confusion within
music and emotion studies stems from the fact that
conceptual frameworks for emotions in music are
applied in an inconsistent manner. This may seem critical and overly pessimistic, as the most common emotion modelsthe discrete and the dimensionalcover
about 70% of the studies. This ostensible consensus also
reveals the curious reliance on a few, contentious models that have been seriously challenged within the emotion literature. Russells (1980) circumplex model, the
emotion model most often used in music and emotion
studies, has been criticized over the years, and models
that distinguish between negative and positive affect
dimensions (Thayer, 1989; Watson et al., 1988) and
thus account for mixed emotions (Larsen & Diener,
1992; Schimmack & Grob, 2000) have not only been
proposed but critically examined for nearly two decades. The latter critique of Russells (1980) model is
particularly relevant to the domain of music, since several studies have now demonstrated that mixed feelings
and perceptions in response to music are common
(Hunter et al., 2008; Hunter, Schellenberg, & Schimmack, 2010), yet impossible to capture by such a model.
Also, the selection principles for the stimuli have not
always been consistent with the type of emotion model
adopted, as there are studies where discrete emotion
excerpts were rated using dimensional concepts (e.g.,
Khalfa, Peretz, Blondin, & Manon, 2002; Kreutz et al.,
2008). Some studies even suggest that the two models
are based on entirely different processes (Dellacherie,
Ehrle, & Samson, 2008; Gosselin et al., 2005; Khalfa,
Delbe et al., 2008), and thus these possible underlying
differences should be considered before freely mixing
the two models. Furthermore, the increasing popularity
of the simultaneous use of discrete and dimensional
models is indicative of a search for an appropriate emotion framework that would encompass both.
Music-specific modelsthough still relatively rarely
usedare clearly going to become more common in the
near future, as a number of influential studies with lists of
music-specific emotion concepts have recently emerged
(Juslin et al., 2008, Juslin & Laukka, 2004; Zentner et al.,
2008). As these studies are not yet fully compatible with
each other, more research is needed to assess how factors
such as musical genre, individual differences, and the

This content downloaded from 129.240.0.83 on Mon, 6 May 2013 08:34:26 AM


All use subject to JSTOR Terms and Conditions

Review of Music and Emotion Studies

various mechanisms through which music induces emotions (for these, see Juslin & Vastfjall, 2008) contribute to
the variations in music-specific emotion concepts.
We strongly recommend that future studies would
more explicitly justify their choices with regard to the
emotion models and concepts usedboth in the selection of the stimuli and in the subsequent self-reports/
analysesand include critical discussions on the applicability and theoretical background of their chosen
emotion models. Finally, the focus on either felt or
perceived emotions needs to be unambiguously qualified in future studies, and care should be taken in
communicating this to the participants as well.
EMOTION VALIDATION USING MULTIPLE APPROACHES

The application of several research approaches in a single


study is not yet that common in the field of music and
emotions, despite the fact that the many separate components that make up emotions may only be satisfactorily studied by using a combination of different
approaches (biological, self-report, observation, etc.).
Currently, the most popular approach is to use selfreport methods. This is probably due to the fact that
nearly all other approaches also use self-reports as
a complementary methodin order to verify that the
participants are actually experiencing the appropriate
emotionswhile the actual focus is on biological, clinical, music analytic, or individual and cultural aspects of
emotions. It is important to stress that although selfreport measures are an indispensable means of obtaining information about the emotions, feelings, and
moods experienced by people, they have certain weaknesses (e.g., demand characteristics and the need for
accurate introspection) that render the results obtained
using these methods inconclusive. On the other hand,
the findings obtained using biological or music analytic
methods (for example) cannot be corroborated without
self-reports. A valid and reliable understanding of musicinduced emotions can only emerge when multiple
approachessuch as self-reports, biological measures,
and indirect measures of felt emotion (for a review, see
Vastfjall, 2008)are simultaneously used to investigate
a given problem. However, we understand that this is
a challenging task, made even more complicated by the
extant variety of emotion models and terminologies, not
to mention the two kinds of emotion loci.
ACKNOWLEDGING THE SIGNIFICANCE OF PARTICIPANT
BACKGROUNDS

Most music and emotion studies have focused on samples of convenience (i.e., student populations from
Western universities) and have drawn music from

325

a fairly narrow selection of musical genres (i.e., classical,


jazz, and film soundtracks). It is safe to assume that the
emotional response to a given piece of music will vary
according age, cultural background, and musical taste.
At the most fundamental level, cross-cultural studies on
music and emotion are sorely needed in order to estimate the degree to which findings in the field can be
generalized across cultures. The rarity of cross-cultural
studies (only five in this review) is to be expected, given
that such endeavours are extremely difficult, and are
becoming ever more so due to the effects of continuing
globalization. Even within these five cross-cultural studies, only one study included a sample population that
was not familiar with Western music (Fritz et al., 2009).
Perhaps it could be suggested that not all cross-cultural
studies need to compare two separate cultures (that have
no contact with each other), but useful tests of generalizability could also be conducted within the different
cultural practices in the West (i.e., between different
musical traditions). It may also be that different musical
genres lead to different spectrums of emotions, as the
functions of music often vary depending on the genre.
ECOLOGICALLY VALID MUSIC

Classical music is still the dominant musical genre used


in music and emotion studies (nearly half of the studies
covered here). This echoes what Juslin and Laukka
(2003) identified as one of the central areas for improvement in the field. That is to say that there is a need to
have, in their words, a greater variety of musical materials (p. 804). The widespread use of classical music has
even led some researchers to conclude that liking of
classical music would be a prerequisite for experiencing
strong, music-induced emotions (Kreutz et al., 2008),
which is clearly a problem related to the stimulus material used. As classical music is predominantly used in
music and emotion studies as well as in mood induction
studies (60-70% of the stimuli, see Vastfjall, 2002), we
strongly recommend that at least the overly popular
repertoire within the genre should be avoided. Furthermore, researchers should try to move beyond the outdated view that classical music would be emotionally
and aesthetically superior to other types of music.
The acknowledgement that music-induced emotions
are personal and depend on individual factors should be
acted upon, and also incorporated into studies either by
asking the participants to bring their own music, or by
providing the participants with a large selection of
music to choose from. Music that is ecologically valid
should somehow be relevant for the people participating
in listening experimentsbe it music that they are
exposed to on a daily basis, or music that they choose

This content downloaded from 129.240.0.83 on Mon, 6 May 2013 08:34:26 AM


All use subject to JSTOR Terms and Conditions

326

Tuomas Eerola & Jonna K. Vuoskoski

to listen to themselves. Much could be learned by investigating the music that people actually use to regulate
their own emotions and moods. This is of course not
possible for all kinds of study designs (e.g., factorial
manipulations of various musical features), but it is
clearly something that could be easily improved, provided that a suitable selection of protocols and acoustic
analyses are established.
Despite the above-mentioned recommendations for
using ecological valid stimulus materials, we do advocate
the systematic manipulation of musical features using
factorial designs. Although this may seem to go against
the requirement for using ecologically valid stimuli, studies using controlled stimuli are still needed in order to
disambiguate musical features from each other in a causal
fashion. Finally, in the name of transparency and the
facilitation of the replication, simulation, and generation
of follow-up studies, we strongly urge researchers to
share their stimulus sets in secure online repositories.

experience sampling studies)have already been suggested by Juslin and Laukka (2003), but are nevertheless
still topical. We might also add that despite the current
efforts to capture the temporal nature of music-induced
emotions with continuous self-report and biological
approaches, this paradigm needs to be enlarged upon
and incorporated into the majority of future studies.
The results obtained in this review do not suggest that
enough has already been done in these directions. The
research on music and emotion has the capacity of
being an area in which progress can be made towards
understanding one of the most fascinating aspects of
human life which is the cause of so much else beyond
musicthe full spectrum of emotional experiences. The
obstacles in the way are outlined by Juslin and Vastfjall
(2008) and this current study. These will hopefully be
noted by future researchers who, by systematically
avoiding them, will be able to take more reliable and
purposeful steps towards understanding the psychological and biological foundations of musical emotions.

CONCLUDING REMARKS

This review has identified several key issues that are


currently holding back advances in the field of music
and emotions. We have provided explicit recommendations, and suggested that future studies should especially aim for theoretical consistency, the use of
multiple approaches simultaneously, the use of ecologically valid stimulus material, and the careful consideration of participant backgrounds. Other points we
would like to raisesuch as the need to venture outside
the laboratory to study emotions in the contexts in
which they naturally occur, and the need to use
a broader array of methods (e.g., diary studies,

Author Note

This work was funded by the European Union (BrainTuning FP6-2004-NEST-PATH-028570) and the Academy of
Finland (Finnish Centre of Excellence in Interdisciplinary
Music Research). We thank Alex Reed for the terminological and language-related improvements.
Correspondence concerning this article should be
addressed to Tuomas Eerola, Finnish Centre of Excellence in Interdisciplinary Music Research, Seminaarinkatu 35, B. O. Box 15, FI-40014 University of Jyvaskyla,
Finland. E-mail: tuomas.eerola@jyu.fi

References
*References marked with an asterisk indicate studies included in
the review.
*A DACHI , M., & T REHUB , S. E. (1998). Childrens expression of
emotion in song. Psychology of Music, 26, 133-153.
*A DACHI , M., & T REHUB , S. E. (2002). Decoding the expressive
intentions in childrens songs. Music Perception, 18, 213-224.
ALBERSNAGEL, F. A. (1988). Velten and musical mood induction
procedures: A comparison with accessibility in thought associations. Behaviour Research and Therapy, 26, 79-96.
*ALFREDSON, B. B., RISBERG, J., HAGBERG, B., & GUSTAFSON, L. (2004).
Right temporal lobe activation when listening to emotionally
significant music. Applied Neuropsychology, 11(3), 161-166.
*A LI , S. O., & P EYNIRCIG OLU , Z. F. (2006). Songs and emotions:
Are lyrics and melodies equal partners? Psychology of Music,
34, 511-534.

*A L TMAN , Y. A., A LYANCHIKOVA , Y. O., G UZIKOV , B. M., &


Z AKHAROVA , L. E. (2000). Estimation of short musical fragments in normal subjects and patients with chronic depression. Human Physiology, 26, 553-557.
*A LTENMU LLER , K., S CHUERMANN , K., L IM , V. K., & P ARLITZ , D.
(2002). Hits to the left, flops to the right: Different emotions
during listening to music are reflected in cortical lateralisation
patterns. Neuropsychologia, 40, 2242-2256.
A SMUS , E. (1985). The development of a multidimensional
instrument for the measurement of affective responses to
music. Psychology of Music, 13, 19-30.
*BALKWILL, L.-L., & THOMPSON, W. F. (1999). A cross-cultural
investigation of the perception of emotion in music:
Psychophysical and cultural cues. Music Perception, 17, 43-64.
*B ALKWILL , L.-L., T HOMPSON , W. F., & M ATSUNAGA , R. (2004).
Recognition of emotion in Japanese, Western, and Hindustani

This content downloaded from 129.240.0.83 on Mon, 6 May 2013 08:34:26 AM


All use subject to JSTOR Terms and Conditions

Review of Music and Emotion Studies

music by Japanese listeners. Japanese Psychological Research,


46, 337-349.
*B ANICH , M. T., S TOLAR , N., H ELLER , W., & G OLDMAN , R. B.
(1992). A deficit in right-hemisphere performance after
induction of a depressed mood. Neuropsychology, 5, 20-27.
B ANSE , R., & S HERER , K. R. (1996). Acoustic profiles in vocal
emotion expression. Journal of Personality and Social
Psychology, 70, 614-636.
*B ARALDI , F., D E P OLI , G., & R ODA` , A. (2006). Communicating
expressive intentions with a single piano note. Journal of New
Music Research, 35(3), 197-210.
B ARRETT , L. F., & R USSELL , J. A. (1999). Structure of current
affect. Current Directions in Psychological Science, 8, 10-14.
B ARRETT , L. F., & W AGER , T. (2006). The structure of emotion:
Evidence from the neuroimaging of emotion. Current
Directions in Psychological Science, 15, 79-85.
*B ARTEL , L. R. (1992). The development of the cognitiveaffective response test-music. Psychomusicology, 11, 15-26.
*B AUMGARTNER , T., E SSLEN , M., & J ANCKE , L. (2006). From
emotion perception to emotion experience: Emotions evoked
by pictures and classical music. International Journal of
Psychophysiology, 60, 34-43.
*B EHRENS , G., & G REEN , S. (1993). The ability to identify
emotional content of solo improvisations performed vocally
and on three different instruments. Psychology of Music, 21,
20-33.
*B EVER , T. G. (1988). A cognitive theory of emotion and aesthetics in music. Psychomusicology, 7, 165-175.
*B IGAND , E., V IEILLARD , S., M ADURELL , F., M AROZEAU , J., &
D ACQUET , A. (2005). Multidimensional scaling of emotional
responses to music: The effect of musical expertise and of the
duration of the excerpts. Cognition and Emotion, 19, 11131139.
*B LOOD , A. J., & Z ATORRE , R. J. (2001). Intensely pleasurable
responses to music correlate with activity in brain regions
implicated in reward and emotion. Proceeding of National
Academy of Sciences, 98, 11818-11823.
*B LOOD , A. J., Z ATORRE , R. J., B ERMUDEZ , P., & E VANS , A. C.
(1999). Emotional responses to pleasant and unpleasant music
correlate with activity in paralimbic brain regions. Nature
Neuroscience, 2, 382-287.
*B OONE , R., & C UNNINGHAM , J. (2001). Childrens expression of
emotional meaning in music through expressive body movement. Journal of Nonverbal Behavior, 25, 21-41.
*B OUHUYS , A. L., B LOEM , G. M., & G ROOTHUIS , T. G. G.
(1995). Induction of depressed and elated mood by music
influences the perception of facial emotional expressions in
healthy subjects. Journal of Affective Disorders, 33, 215-226.
B RESIN , R. (2005). What is the color of that music performance.
In Proceedings of the International Computer Music Conference
(pp. 367-370). Barcelona, Spain: International Computer
Music Association.

327

*B RESIN , R., & F RIBERG , A. (2000). Emotional coloring of


computer-controlled music performances. Computer Music
Journal, 24(4), 44-63.
*B RITTIN , R. V., & D UKE , R. A. (1997). Continuous versus
summative evaluations of musical intensity: A comparison of
two methods for measuring overall effect. Journal of Research
in Music Education, 45(2), 245-258.
*B ROSGOLE , L., & W EISMAN , J. (1995). Mood recognition across
the ages. International Journal of Neuroscience, 82, 169-189.
*B ROWN , S., M ARTINEZ , M. J., & P ARSONS , L. M. (2004). Passive
music listening spontaneously engages limbic and paralimbic
systems. NeuroReport, 15, 2033-2037.
B URKHARDT , F., P AESCHKE , A., R OLFES , M., S ENDLMEIER , W., &
W EISS , B. (2005). A database of German Emotional Speech. In
Proceedings of the INTERSPEECH 2005(pp. 1517-1520).
Lisbon, Portugal: ISCA.
C AMURRI , A., L AGERLO F , I., & V OLPE , G. (2003). Recognizing
emotion from dance movement: Comparison of spectator
recognition and automated techniques. International Journal
of Human-Computer Studies, 59, 213-225.
C ANNAM , C., L ANDONE , C., S ANDLER , M., & B ELLO , J. (2006).
The sonic visualiser: A visualisation platform for semantic
descriptors from musical signals. In Proceedings of ISMIR
2006, 7th International Conference on Music Information
Retrieval (pp. 324-327). Victoria, Canada: University of
Victoria, Canada.
*C HAPADOS , C., & L EVITIN , D. (2008). Cross-modal interactions
in the experience of musical performances: Physiological correlates. Cognition, 108, 639-651.
*CHEN, J., YUAN, J., HUANG, H., CHEN, C., & LI, H. (2008).
Music-induced mood modulates the strength of emotional
negativity bias: An ERP study. Neuroscience Letters, 445, 135139.
*C HEN , L., Z HOU , S., & B RYANT , J. (2007). Temporal changes in
mood repair through music consumption: Effects of mood,
mood salience, and individual differences. Media Psychology,
9, 695-713.
C LARK , D. M., & T EASDALE , J. D. (1985). Constraints on the
effect of mood on memory. Journal of Personality and Social
Psychology, 48, 1595-1608.
*C OFFMAN , D. D., G FELLER , K., & E CKERT , M. (1995). Effect of
textual setting, training, and gender on emotional response to
verbal and musical information. Psychomusicology, 14, 117-136.
*C OLLIER , G. L. (2007). Beyond valence and activity in the emotional connotations of music. Psychology of Music, 35, 110-131.
*C OLLIER , W., & H UBBARD , T. (2001a). Judgments of happiness,
brightness, speed, and tempo change of auditory stimuli
varying in pitch and tempo. Psychomusicology, 17, 36-55.
*C OLLIER , W., & H UBBARD , T. (2001b). Musical scales and
evaluations of happiness and awkwardness: Effects of pitch,
direction, and scale mode. American Journal of Psychology,
114, 355-375.

This content downloaded from 129.240.0.83 on Mon, 6 May 2013 08:34:26 AM


All use subject to JSTOR Terms and Conditions

328

Tuomas Eerola & Jonna K. Vuoskoski

*C OOK , N. (2007). The sound symbolism of major and minor


harmonies. Music Perception, 24, 315319.
*C OSTA , M., B ITTI , P. E. R., & B ONFIGLIOLI , L. (2000).
Psychological connotations of harmonic musical intervals.
Psychology of Music, 28, 4-22.
*C OSTA , M., F INE , P. A., & B ITTI , P. E. R. (2004). Interval
distributions, mode, and tonal strength of melodies as predictors of perceived emotion. Music Perception, 22, 1-14.
COWIE, R., & CORNELIUS, R. (2003). Describing the emotional states
that are expressed in speech. Speech Communication, 40, 5-32.
*C UNNINGHAM , J., & S TERLING , R. (1988). Developmental
change in the understanding of affective meaning in music.
Motivation and Emotion, 12, 399-413.
*D AHL , S., & F RIBERG , A. (2007). Visual perception of expressiveness in musicians body movements. Music Perception, 24,
433-454.
*D ALLA B ELLA , S., P ERETZ , I., R OUSSEAU , L., & G OSSELIN , N.
(2001). A developmental study of the affective value of tempo
and mode in music. Cognition, 80, B1-10.
*D ELLACHERIE , D., E HRLE , N., & S AMSON , S. (2008). Is the
neutral condition relevant to study musical emotion in
patients? Music Perception, 25, 285-294.
*D IBBEN , N. (2004). The role of peripheral feedback in emotional
experience with music. Music Perception, 22, 79-115.
*D IBBEN , N. (2006). Subjectivity and the construction of emotion
in the music of Bjork. Music Analysis, 25, 171-197.
D OUGLAS -C OWIE , E, C AMPBELL , N, C OWIE , R & R OACH , P.
(2003). Emotional speech: Towards a new generation of
databases. Speech Communication, 40, 70-75.
E EROLA , T., & V UOSKOSKI , J. K. (2011). A comparison of the
discrete and dimensional models of emotion in music.
Psychology of Music, 39, 18-49.
*E FIMOVA , I., & B UDYKA , E. (2006). The use of emotional evaluations of short musical fragments for analyzing the functional states of students. Human Physiology, 32, 278-281.
E IJCK , K. (2001). Social differentiation in musical taste patterns.
Social Forces, 79, 1163-1118.
E KMAN , P. (1992). An argument for basic emotions. Cognition
and Emotion, 6, 169-200.
E LFENBEIN , H. A., & A MBADY , N. (2002). On the universality
and cultural specificity of emotion recognition: A metaanalysis. Psychological Bulletin, 128, 203-235.
*E LLIS , R., & S IMONS , R. (2005). The impact of music on subjective and physiological indices of emotion while viewing
films. Psychomusicology, 19, 15-40.
*E SCHRICH , S., M U NTE , T., & A LTENMU LLER , E. (2008).
Unforgettable film music: The role of emotion in episodic
long-term memory for music. BMC Neuroscience, 9(48), doi:
10.1186/1471-2202-9-48
*E RKKILA , J., G OLD , C., F ACHNER , J., A LA -R UONA , E.,
P UNKANEN , M., & V ANHALA , M. (2008). The effect of
improvisational music therapy on the treatment of depression:

Protocol for a randomised controlled trial. BMC Psychiatry, 8,


50, doi: 10.1186/1471-244X-8-50
*E TZEL , J. A., J OHNSEN , E. L., D ICKERSON , J., T RANEL , D., &
A DOLPHS , R. (2006). Cardiovascular and respiratory responses
during musical mood induction. International Journal of
Psychophysiology, 61, 57-69.
*E VANS , P., & S CHUBERT , E. (2008). Relationships between
expressed and felt emotions in music. Musicae Scientiae, 12, 75-99.
*F AITH , M., & T HAYER , J. F. (2001). A dynamical systems
interpretation of a dimensional model of emotion.
Scandinavian Journal of Psychology, 42, 121-133.
F ARNSWORTH , P. R. (1954). A study of the Hevner adjective list.
Journal of Aesthetics and Art Criticism, 13, 97-103.
*F LOM , R., G ENTILE , D., & P ICK , A. (2008). Infants discrimination of happy and sad music. Infant Behavior and
Development, 31, 716-728.
*F LORES -G UTIE RREZ , E. O., DI AZ , J.-L., B ARRIOS , F. A., F AVILA ., RI O -P ORTILLA , Y., & C ORSI H UMARA , R., G UEVARA , M. A
C ABRERA , M. (2007). Metabolic and electric brain patterns
during pleasant and unpleasant emotions induced by music
masterpieces. International Journal of Psychophysiology, 65,
69-84.
*F REDRICKSON , W. E. (1995). A comparison of perceived musical
tension and aesthetic response. Psychology of Music, 23, 81-87.
*F REGO , R. J. D. (1999). Effects of aural and visual conditions on
response to perceived artistic tension in music and dance.
Journal of Research in Music Education, 47, 31-43.
F RIJDA , N. H. (2007). What might an emotion be? Comments on
the comments. Social Science Information, 46, 433-443.
*F RITZ , T., J ENTSCHKE , S., G OSSELIN , N., S AMMLER , D., P ERETZ ,
I., T URNER , R., ET AL . (2009). Universal recognition of three
basic emotions in music. Current Biology, 19, 573-576.
*G ABRIELSSON , A. (2002). Emotion perceived and emotion felt:
Same or different. Musicae Scientiae, 2001-2002, 123-147.
*G ABRIELSSON , A., & J USLIN , P. N. (1996). Emotional expression
in music performance: Between the performers intention and
the listeners experience. Psychology of Music, 24, 68-91.
*G ABRIELSSON , A., & L INDSTRO M , E. (1995). Emotional expression in synthesizer and sentograph performance.
Psychomusicology, 14, 94-116.
G ABRIELSSON , A., & L INDSTRO M , E. (2001). The influence of
musical structure on emotional expression. In P. N. Juslin & J.
A. Sloboda (Eds.), Music and emotion: Theory and research
(pp. 235-239). Oxford, UK: Oxford University Press.
G ABRIELSSON , A., & L INDSTRO M , E. (2010). The influence of
musical structure on emotional expression. In P. N. Juslin & J.
A. Sloboda (Eds.), Handbook of music and emotion: Theory,
research, application (pp. 367-400). New York: Oxford
University Press.
*G ABRIELSSON , A., & L INDSTRO M W IK , S. (2003). Strong
experiences related to music: A descriptive system. Musicae
Scientiae, 7, 157-217.

This content downloaded from 129.240.0.83 on Mon, 6 May 2013 08:34:26 AM


All use subject to JSTOR Terms and Conditions

Review of Music and Emotion Studies

*G AGNON , L., & P ERETZ , I. (2000). Laterality effects in processing tonal and atonal melodies with affective and nonaffective
task instructions. Brain and Cognition, 43, 206-10.
*G AGNON , L., & P ERETZ , I. (2003). Mode and tempo relative
contributions to "happy - sad" judgments in equitone melodies.
Cognition and Emotion, 17, 25-40.
*G ERARDI , G., & G ERKEN , L. (1995). The development of affective responses to modality and melodic contour. Music
Perception, 12, 279-279.
*G ERINGER , J. M., C ASSIDY , J. W., & B YO , J. L. (1996). Effects of
music with video on responses of nonmusic majors: An
exploratory study. Journal of Research in Music Education, 44,
240-251.
*G FELLER , K., A SMUS , E. P., & E CKERT , M. (1991). An investigation of emotional response to music and text. Psychology of
Music, 19, 128-141.
*G FELLER , K., & C OFFMAN , D. D. (1991). An investigation of
emotional response of trained musicians to verbal and music
information. Psychomusicology, 10, 31-48.
*G ILBOA , A., B ODNER , E., & A MIR , D. (2006). Emotional communicability in improvised music: The case of music therapists. Journal of Music Therapy, 43, 198-225.
*G IOMO , C. J. (1993). An experimental study of childrens
sensitivity to mood in music. Psychology of Music, 21,
141-162.
*G LOECKNER , N. (2006). Psychophysiological investigation of
emotional states evoked by music. In M. Baroni, A. R. Addessi,
R. Caterina, & M. Costa (Eds.), Proceedings of the 9th
International Conference on Music Perception and Cognition
(pp. 95-106). Bologna, Italy: ICMPC-ESCOM.
*G OMEZ , P., & D ANUSER , B. (2004). Affective and physiological
responses to environmental noises and music. International
Journal of Psychophysiology, 53, 91-103.
*G OMEZ , P., & D ANUSER , B. (2007). Relationships between
musical structure and psychophysiological measures of emotion. Emotion, 7, 377-387.
*G ORN , G., P HAM , M. T., & S IN , L. Y. (2001). When arousal
influences ad evaluation and valence does not (and vice versa).
Journal of Consumer Psychology, 11, 43-55.
*G OSSELIN , N., P ERETZ , I., J OHNSEN , E., & A DOLPHS , R. (2007).
Amygdala damage impairs emotion recognition from music.
Neuropsychologia, 45, 236-244.
*G OSSELIN , N., P ERETZ , I., N OULHIANE , M., H ASBOUN , D.,
B ECKETT , C., B AULAC , M., & S AMSON , S. (2005). Impaired
recognition of scary music following unilateral temporal lobe
excision. Brain, 128, 628-640.
*G OSSELIN , N., S AMSON , S., A DOLPHS , R., N OULHIANE , M., R OY ,
M., H ASBOUN , D., ET AL . (2006). Emotional responses to
unpleasant music correlates with damage to the parahippocampal cortex. Brain, 129, 2585-2592.
*G OYDKE , K. N., A LTENMU LLER , E., M OLLER , J., & M UNTE , T. F.
(2004). Changes in emotional tone and instrumental timbre

329

are reflected by the mismatch negativity. Cognitive Brain


Research, 21, 351-359.
*G REEN , A., B RENTSEN , K., S TDKILDE -J RGENSEN , H.,
W ALLENTIN , M., R OEPSTORFF , A., & V UUST , P. (2008). Music
in minor activates limbic structures: A relationship with
dissonance? Neuroreport, 19, 711-715.
*GREGORY, A. H., & VARNEY, N. (1996). Cross-cultural comparisons
in the affective response to music. Psychology of Music, 24,
47-52.
*G REWE , O., N AGEL , F., K OPIEZ , R., & A LTENMU LLER , E.
(2007a). Emotions over time: Synchronicity and development
of subjective, physiological, and facial affective reactions to
music. Emotion, 7, 774-788.
*G REWE , O., N AGEL , F., K OPIEZ , R., & A LTENMU LLER , E.
(2007b). Listening to music as a re-creative process:
Physiological, psychological, and psychoacoustical correlates
of chills and strong emotions. Music Perception, 24, 297-314.
*G UPTA , U., & G UPTA , B. (2005). Psychophysiological responsivity to indian instrumental music. Psychology of Music, 33,
363-372.
*H ARGREAVES , D. J., & N ORTH , A. C. (1999). The functions of
music in everyday life: Redefining the social in music psychology. Psychology of Music, 27, 71-83.
*H ERZ , R. S. (1998). An examination of objective and subjective
measures of experience associated to odors, music and paintings. Empirical Studies of the Arts, 16, 137-152.
H EVNER , K. (1936). Experimental studies of the elements of
expression in music. American Journal of Psychology, 48,
248-268.
*H OLBROOK , M. B., & A NAND , P. (1992). The effects of situation,
sequence, and features on perceptual and affective responses to
product designs: The case of aesthetic consumption. Empirical
Studies of the Arts, 10, 19-31.
*HOSHINO, E. (1996). The feeling of musical mode and its emotional character in a melody. Psychology of Music, 24, 29-46.
*H UNTER , P. G., S CHELLENBERG , G. E., & S CHIMMACK , U.
(2008). Mixed affective responses to music with conflicting
cues. Cognition and Emotion, 22, 327-352.
H UNTER , P. G., S CHELLENBERG , E. G., & S CHIMMACK , U. (2010).
Feelings and perceptions of happiness and sadness induced by
music: Similarities, differences, and mixed emotions.
Psychology of Aesthetics, Creativity, and the Arts, 4, 47-56.
*H URON , D. (2008). A comparison of average pitch height and
interval size in major-and minor-key themes: Evidence consistent with affect-related pitch prosody. Empirical Musicology
Review, 3, 59-63.
*H USAIN , G., T HOMPSON , W. F., & S CHELLENBERG , E. G. (2002).
Effects of musical tempo and mode on arousal, mood, and
spatial abilities. Music Perception, 20, 151-171.
*I LIE , G., & T HOMPSON , W. F. (2006). A comparison of acoustic
cues in music and speech for three dimensions of affect. Music
Perception, 23, 319-329.

This content downloaded from 129.240.0.83 on Mon, 6 May 2013 08:34:26 AM


All use subject to JSTOR Terms and Conditions

330

Tuomas Eerola & Jonna K. Vuoskoski

*I WAKI , H. T., H AYASHI , M., & H ORI , T. (1997). Changes in


alpha band EEG activity in the frontal area after stimulation
with music of different affective content. Perceptual and Motor
Skills, 84, 515-526.
*I WANAGA , M., & T SUKAMOTO , M. (1998). Preference for
musical tempo involving systematic variations of presented
tempi for known and unknown musical excerpts. Perceptual
and Motor Skills, 86, 31-41.
I ZARD , C. E. (2007). Basic emotions, natural kinds, emotion
schemas, and a new paradigm. Perspectives on Psychological
Science, 2, 260-280.
*J USLIN , P. N. (1997a). Can results from studies of perceived
expression in musical performances be generalized across
response formats? Psychomusicology, 16, 77-101.
*J USLIN , P. N. (1997b). Emotional communication in music
performance: A functionalist perspective and some data. Music
Perception, 14, 383-418.
*J USLIN , P. N. (1997c). Perceived emotional expression in
synthesized performances of a short melody: Capturing
the listenerss judgement policy. Musicae Scientiae, 1,
225-256.
*J USLIN , P. N. (2000). Cue utilization in communication of
emotion in music performance: Relating performance to
perception. Journal of Experimental Psychology: Human
Perception and Performance, 26, 1797-1813.
*J USLIN , P. N., F RIBERG , A., & B RESIN , R. (2001). Toward
a computational model of expression in music performance:
The GERM model. Musicae Scientiae, 6, 63-122.
*J USLIN , P., & L AUKKA , P. (2000). Improving emotional communication in music performance through cognitive feedback.
Musicae Scientiae, 4, 151-184.
*JUSLIN, P. N., & LAUKKA, P. (2003). Communication of
emotions in vocal expression and music performance:
Different channels, same code? Psychological Bulletin, 129,
770-814.
*J USLIN , P. N., & L AUKKA , P. (2004). Expression, perception,
and induction of musical emotions: A review and a questionnaire study of everyday listening. Journal of New Music
Research, 33, 217-238.
*J USLIN , P. N., L ILJESTRO M , S., V A STFJA LL , D., B ARRADAS , G., &
S ILVA , A. (2008). An experience sampling study of emotional
reactions to music: Listener, music, and situation. Emotion, 8,
668-683.
*J USLIN , P. N., & M ADISON , G. (1999). The role of timing patterns in recognition of emotional expression from musical
performance. Music Perception, 17, 197-221.
J USLIN , P. N., & S LOBODA , J. A. (Eds.). (2001). Music and
emotion: Theory and research. Oxford, UK: Oxford University
Press.
J USLIN , P. N., & S LOBODA , J. A. (Eds.). (2010). Handbook of
music and emotion: Theory, research and applications. Oxford,
UK: Oxford University Press.

*J USLIN , P. N., & V A STFJA LL , D. (2008). Emotional responses to


music: The need to consider underlying mechanisms.
Behavioral and Brain Sciences, 31, 559-575.
*JUSLIN, P. N., & ZENTNER, M. R. (2001). Current trends in the
study of music and emotion: Overture. Musicae Scientiae, 6,
3-22.
*KABUTO, M., KAGEYAMA, T., & NITTA, H. (1993). EEG power
spectrum changes due to listening to pleasant music and their
relation to relaxation effects. Nippon Eiseigaku Zasshi, 48,
807-818.
*K ALLINEN , K. (2005). Emotional ratings of music excerpts in the
Western art music repertoire and their self-organization in the
Kohonen neural network. Psychology of Music, 33, 373-393.
*K ALLINEN , K., & R AVAJA , N. (2004). Emotion-related effects of
speech rate and rising vs. falling background music melody
during audio news: The moderating influence of personality.
Personality and Individual Differences, 37, 275288.
*KALLINEN, K., & RAVAJA, N. (2006). Emotion perceived and
emotion felt: Same and different. Musicae Scientiae, 10,
191-213.
*K AMENETSKY , S. B., H ILL , D. S., & T REHUB , S. E. (1997). Effect
of tempo and dynamics on the perception of emotion in music.
Psychology of Music, 25, 149-160.
*K AMINSKA , Z., & W OOLF , J. (2000). Melodic line and emotion:
Cookes theory revisited. Psychology of Music, 28, 133-153.
*K ASTNER , M. P., & C ROWDER , R. G. (1990). Perception of the
major/minor distinction: Emotional connotations in young
children. Music Perception, 8, 189-202.
*K HALFA , S., D ELBE , C., B IGAND , E., R EYNAUD , E., C HAUVEL , P.,
& L IE GEOIS -C HAUVEL , C. (2008). Positive and negative music
recognition reveals a specialization of mesio-temporal structures in epileptic patients. Music Perception, 25, 295-302.
*K HALFA , S., P ERETZ , I., B LONDIN . J.-P., & M ANON , R. (2002).
Event-related skin conductance responses to musical emotions
in humans. Neuroscience Letters, 328, 145-149.
*K HALFA , S., R OY , M., R AINVILLE , P., D ALLA B ELLA , S., &
P ERETZ , I. (2008). Role of tempo entrainment in psychophysiological differentiation of happy and sad music?
International Journal of Psychophysiology, 68, 17-26.
*K HALFA , S., S CHON , D., A NTON , J., & L IE GEOIS -C HAUVEL , C.
(2005). Brain regions involved in the recognition of happiness
and sadness in music. Neuroreport, 16, 1981-1984.
*K IM , J., & A NDRE , E. (2008). Emotion recognition based on
physiological changes in music listening. IEEE Transactions
on Pattern Analysis and Machine Intelligence, 30,
2067-2083.
*K INSELLA , M. P., & J ONES , V. (1990). Judgement of mood in
music following right hemisphere damage. Archives of Clinical
Neuropsychology, 5, 359-371.
KIVY, P. (1990). Music alone: Philosophical reflections on the
purely musical experience. Ithaca, NY: Cornell University
Press.

This content downloaded from 129.240.0.83 on Mon, 6 May 2013 08:34:26 AM


All use subject to JSTOR Terms and Conditions

Review of Music and Emotion Studies

*K OELSCH , S. (2005). Investigating emotion with music:


Neuroscientific approaches. Annals of the New York Academy
of Sciences, 1060, 412-418.
*K OELSCH , S., F RITZ , T., C RAMON , D. Y., M U LLER , K., &
F RIEDERICI , A. D. (2006). Investigating emotion with music:
An fMRI study. Human Brain Mapping, 27, 239-250.
*K OELSCH , S., F RITZ , T., & S CHLAUG , G. (2008). Amygdala
activity can be modulated by unexpected chord functions
during music listening. Neuroreport, 19, 1815-1819.
*K OELSCH , S., K ILCHES , S., S TEINBEIS , N., & S CHELINSKI , S.
(2008). Effects of unexpected chords and of performers
expression on brain responses and electrodermal activity. PLoS
ONE, 3(7), e2631.
*K ONEC NI , V. J. (2008). Does music induce emotion? A theoretical and methodological analysis. Psychology of Aesthetics,
Creativity, and the Arts, 2, 115-129.
*K ONEC NI , V. J., B ROWN , A., & W ANIC , R. A. (2008).
Comparative effects of music and recalled life-events on
emotional state. Psychology of Music, 36, 289-308.
*KONECNI, V. J., WANIC, R. A., & BROWN, A. (2007). Emotional
and aesthetic antecedents and consequences of music-induced
thrills. American Journal of Psychology, 120, 619-643.
*K OPIEZ , R. (2007). EMuJoy: Software for continuous measurement of perceived emotions in music. Behavior Research
Methods, 39, 283-290.
*KORHONEN, M. D., CLAUSI, D. A., & JERNIGAN, M. E. (2006).
Modeling emotional content of music using system identification. IEEE Transactions on Systems, Man, and Cybernetics,
36, 588-599.
*K REUTZ , G., B ONGARD , S., R OHRMANN , S., H ODAPP , V., &
G REBE , D. (2004). Effects of choir singing or listening on
secretory immunoglobulin a, cortisol, and emotional state.
Journal of Behavioral Medicine, 27, 623-635.
*KREUTZ, G., BONGARD, S., & VON JUSSIS, J. (2002).
Cardiovascular responses to music listening: The effects of
musical expertise and emotional expression. Musicae Scientiae,
6, 257-278.
*K REUTZ , G., O TT , U., T EICHMANN , D., O SAWA , P., & V AITL ,
D. (2008). Using music to induce emotions: Influences of
musical preference and absorption. Psychology of Music, 36,
101-126.
*K RUMHANSL , C. L. (1997). An exploratory study of musical
emotions and psychophysiology. Canadian Journal of
Experimental Psychology, 51, 336-352.
*K RUMHANSL , C. L. (1998). Topic in music: An empirical study
of memorability, openness and emotion in Mozarts string
quintet in C major and Beethovens string quartet in A minor.
Music Perception, 16, 119-134.
L ANG , P. J., B RADLEY , M. M., & C UTHBERT , B. N. (1999).
International affective picture system IAPS: Technical manual
and affective ratings. Gainesville, FL: The Center for Research
in Psychophysiology, University of Florida.

331

L ARSEN , R., & D IENER , E. (1992). Promises and problems with


the circumplex model of emotion. Review of personality and
social psychology, 13, 25-59.
L ARTILLOT , O., & T OIVIAINEN , P. (2007). MIR in Matlab (II), A
toolbox for musical feature extraction from audio. In S. Dixon,
D. Bainbridge, & R. Typke (Eds.), Proceedings of the 8th
International Conference on Music Information Retrieval (pp.
237-244). Vienna, Austria: Osterreichische Computer
Gesellschaft.
*LAUKKA, P. (2007). Uses of music and psychological well-being
among the elderly. Journal of Happiness Studies, 8, 215-241.
*L AUKKA , P., & G ABRIELSSON , A. (2000). Emotional expression
in drumming performance. Psychology of Music, 28, 181-189.
*L EHMANN , A. C. (1997). Research note: Affective responses to
everyday life events and music listening. Psychology of Music,
25, 84-90.
*L EMAN , M., V ERMEULEN , V., D E V OOGDT , L., M OELANTS , D.,
& L ESAFFRE , M. (2005). Prediction of musical affect using a combination of acoustic structural cues. Journal of New Music
Research, 34, 39-67.
L ERDAHL , F. (2001). Tonal pitch space. New York: Oxford
University Press.
L ESIUK , T. (2005). The effect of music listening on work performance the effect of music listening on work performance
the effect of music listening on work performance. Psychology
of Music, 33, 173-191.
*LINDSTROM, E. (2003). The contribution of immanent and
performed accents to emotional expression in short tone
sequences. Journal of New Music Research, 32, 269-280.
*L INDSTRO M , E. (2006). Impact of melodic organization on
perceived structure and emotional expression in music.
Musicae Scientiae, 10, 85-117.
*L INDSTRO M , E., J USLIN , P. N., B RESIN , R., & W ILLIAMON , A.
(2003). Expressivity comes from within your soul: A questionnaire study of music students perspectives on expressivity.
Research Studies in Music Education, 20, 23-47.
*L IVINGSTONE , S., M UHLBERGER , R., B ROWN , A., & L OCH , A.
(2007). Controlling musical emotionality: An affective computational architecture for influencing musical emotions.
Digital Creativity, 18, 43-53.
*L ONDON , J. (2001). Some theories of emotion in music and their
implications for research in music psychology. Musicae
Scientiae, 6, 23-36.
*L OWIS , M. J. (2002). Music as a trigger for peak experiences
among a college staff population. Creativity Research Journal,
14, 351-359.
*L U , L., L IU , D., & Z HANG , H. (2006). Automatic mood detection and tracking of music audio signals. IEEE Transactions on
Audio, Speech, and Language processing, 14, 5-18.
*LUCK, G., RIIKKILA, K., LARTILLOT, O., ERKKILA, J., TOIVIAINEN, P.,
MAKELA, A., ET AL. (2006). Exploring relationships between
level of mental retardation and features of music therapy

This content downloaded from 129.240.0.83 on Mon, 6 May 2013 08:34:26 AM


All use subject to JSTOR Terms and Conditions

332

Tuomas Eerola & Jonna K. Vuoskoski

improvisations: A computational approach. Nordic Journal of


Music Therapy, 15, 30-48.
*L UCK , G., R IIKKILA , K., L ARTILLOT , O., E RKKILA , J.,
T OIVIAINEN , P., M A KELA , A., ET AL . (2008). Modelling the
relationships between emotional responses to, and musical
content of, music therapy improvisations. Psychology of Music,
36, 25-45.
*L UNDQVIST , L. G., C ARLSSON , F., & H ILMERSSON , P., &
J USLIN , P. N. (2009). Emotional responses to music:
Experience, expression, and physiology. Psychology of
Music, 37, 61-90.
M AC D ONALD , R. A. R., M ITCHELL , L. A., D ILLON , T., S ERPELL ,
M. G., D AVIES , J. B., & A SHLEY , E. A. (2003). An empirical
investigation of the anxiolytic and pain reducing effects of
music. Psychology of Music, 31, 187-203.
*M AC D ORMAN , K., O UGH , S., & H O , C. (2007). Automatic
emotion prediction of song excerpts: Index construction,
algorithm design, and empirical comparison. Journal of New
Music Research, 36, 281-299.
*M ADISON , G. (2000). Properties of expressive variability patterns in music performances. Journal of New Music Research,
29, 335-356.
*M ADSEN , C. K. (1997a). Emotional response to music as measured by the two-dimensional CRDI. Journal of Music
Therapy, 34, 187-199.
*M ADSEN , C. K. (1997b). Emotional response to music.
Psychomusicology, 16, 59-67.
*M ADSEN , C. K. (1998). Emotion versus tension in Haydns
symphony No. 104 as measured by the two-dimensional
continuous response digital interface. Journal of Research in
Music Education, 46, 546-554.
*M ADSEN , C. K., & F REDERICKSON , W. E. (1993). The experience
of musical tension: A replication of Nielsens research using
the continuous response digital interface. Journal of Music
Therapy, 30, 46-63.
*M AKRIS , I. (2003). Judging the pleasantness of contour-rhythmpitch-timbre musical combinations. American Journal of
Psychology, 116, 581-611.
*M ANCINI , M., B RESIN , R., & P ELACHAUD , C. (2007). A virtual
head driven by music expressivity. IEEE Transactions on Audio
Speech and Language Processing, 15, 1833-1845.
*M ARTIN , M. A., & M ETHA , A. (1997). Recall of early childhood
memories through musical mood induction. Arts in
Psychotherapy, 24, 447-454.
*MATSUURA, M. (1998). Effects of background music on differential digital photoplethysmograms and mood states. Japanese
Journal of Physiological Psychology and Psychophysiology, 16,
13-23.
M AUSS , I., & R OBINSON , M. (2009). Measures of emotion: A
review. Cognition and Emotion, 23, 209-237.
*M C A DAMS , S., V INES , B., V IEILLARD , S., S MITH , B., &
R EYNOLDS , R. (2004). Influences of large-scale form on

continuous ratings in response to a contemporary piece in


a live concert setting. Music Perception, 22, 297-350.
*M C E VILLY , D. K. (1999). Chills and tempo. Music Perception,
16, 457-462.
*M C F ARLAND , R. A., & K ENNISON , R. F. (1989a). Handedness
affects emotional valence asymmetry. Perceptual and Motor
Skills, 68, 435-441.
*M C F ARLAND , R. A., & K ENNISON , R. F. (1989b ). Asymmetry in
the relationship between finger temperature changes and
emotional state in males. Biofeedback and Self Regulation, 14,
281-290.
M C N AIR , D. M., L ORR , M., & D ROPPLEMAN , L. F. (1981).
Manual for the profile of mood states. San Diego, CA:
Educational and Industrial Testing Service.
M EYER , L. B. (1956). Emotion and meaning in music. Chicago,
IL: Chicago University Press.
*M EYER , R. K., P ALMER , C., & M AZO , M. (1998). Affective and
coherence responses to Russian laments. Music Perception, 16,
135-150.
*M ILLER , M. M., & S TRONGMAN , K. T. (2002). The emotional
effects of music on religious experience: A study of the
pentecostal-charismatic style of music and worship. Psychology
of Music, 30, 8-27.
*M ITTERSCHIFFTHALER , M. T., F U , C. H. Y., D ALTON , J. A.,
A NDREW , C. M., & W ILLIAMS , S. C. R. (2007). A functional
MRI study of happy and sad affective states induced by classical music. Human Brain Mapping, 28, 1150-1162.
*M OROZOV , V. (1996). Emotional expressiveness of the singing
voice: The role of macrostructural and microstructural modifications of spectra. Logopedics Phonatrics Vocology, 21, 49-58.
*M ORTON B. J., & T REHUB , S. E. (2007). Childrens judgements
of emotion in song. Psychology of Music, 35, 629-639.
M ULDER , J., T ER B OGT , T. F. M., R AAIJMAKERS , Q. A. W.,
G ABHAINN , S. N., & S IKKEMA , P. (2009). From death metal to
R&B? Consistency of music preferences among Dutch adolescents and young adults. Psychology of Music, 38, 67-83.
*N AGEL , F., K OPIEZ , R., G REWE , O., & A LTENMU LLER , E. (2008).
Psychoacoustical correlates of musically induced chills.
Musicae Scientiae, 12, 101-113.
*N APOLES , J., & M ADSEN , C. (2008). Measuring emotional
responses to music within a classroom setting. International
Journal of Music Education, 26, 63-71.
*N ATER , U. M., A BBRUZZESE , E., K REBS , M., & E HLERT , U.
(2006). Sex differences in emotional and psychophysiological
responses to musical stimuli. International Journal of
Psychophysiology, 62, 300-308.
*N AWROT , E. S. (2003). The perception of emotional expression
in music: Evidence from infants, children and adults.
Psychology of Music, 31, 75-92.
HMAN , R. (1993). On perception
*N IELZEN , S., O LSSON , O., & O
of complex sound in schizophrenia and mania.
Psychopathology, 26, 13-23.

This content downloaded from 129.240.0.83 on Mon, 6 May 2013 08:34:26 AM


All use subject to JSTOR Terms and Conditions

Review of Music and Emotion Studies

*N ORTH , A. C., & H ARGREAVES , D. J. (1997). Liking, arousal


potential, and the emotions expressed by music. Scandinavian
Journal of Psychology, 38, 45-53.
N ORTH , A. C., H ARGREAVES , D. J., & H ARGREAVES , J. J. (2004).
The uses of music in everyday life. Music Perception, 22, 63-99.
N USSBAUM , C. O. (2007). The musical representation: Meaning,
ontology, and emotion. Boston, MA; MIT Press.
*N YKLI C EK , I., T HAYER , J. F., & V AN D OORNEN , L. J. P. (1997).
Cardiorespiratory differentiation of musically-induced emotions. Journal of Psychophysiology, 11, 304-321.
*O CKELFORD , A. (2005). Relating musical structure and content
to aesthetic response: A model and analysis of Beethovens
Piano Sonata Op. 110. Journal of the Royal Musical
Association, 130, 74-118.
O RTONY , A., & T URNER , T. J. (1990). Whats basic about basic
emotions? Psychological Review, 97, 315-331.
*P ALLESEN , K., B RATTICO , E., B AILEY , C., K ORVENOJA , A.,
K OIVISTO , J., G JEDDE , A., & C ARLSON , S. (2005). Emotion
processing of major, minor, and dissonant chords. Annals of
the New York Academy of Sciences, 1060, 450-453.
*P ANKSEPP , J., & B EKKEDAL , M. Y. V. (1997). The affective
cerebral consequence of music: Happy vs. sad effects on the
EEG and clinical implications. International Journal of Arts
Medicine, 5, 18-27.
*P ANKSEPP , J. (1992). A critical role for affective neuroscience in
resolving what is basic about basic emotions. Psychological
Review, 99, 554-560.
*P ANKSEPP , J. (1995). The emotional sources of "chills" induced
by music. Music Perception, 13, 171-207.
P ANKSEPP , J. (1998). Affective neuroscience: The foundations of
human and animal emotions. Oxford, UK: Oxford University
Press.
P ELLETIER , C. L. (2004). The effect of music on decreasing
arousal due to stress: A meta-analysis. Journal of Music
Therapy, 41, 192-214.
P ERETZ , I. (2010). Towards a neurobiology of musical emotions.
In P. N. Juslin & J. A. Sloboda (Eds.), Handbook of music and
emotion: Theory, research, applications (pp. 99-126). Oxford,
UK: Oxford University Press.
*P ERETZ , I., B LOOD , A. J., P ENHUNE , V., & Z ATORRE , R. (2001).
Cortical deafness to dissonance. Brain, 124, 928-940.
*PERETZ, I., & GAGNON, L. (1999). Dissociation between recognition and emotional judgements for melodies. Neurocase, 5,
21-30.
*P ERETZ , I., G AGNON , L., & B OUCHARD , B. (1998). Music and
emotion: Perceptual determinants, immediacy, and isolation
after brain damage. Cognition, 68, 111-141.
*P IGNIATELLO , M. F., & R ASAR , L. A. (1989). A psychophysiological comparison of the Velten and musical mood induction
techniques. Journal of Music Therapy, 26, 140-154.
P OSNER , J., R USSELL , J. A., & P ETERSON , B. S. (2005). The circumplex model of affect: An integrative approach to affective

333

neuroscience, cognitive development, and psychopathology.


Development and Psychopathology, 17, 715-734.
*R AVAJA , N., & K ALLINEN , K. (2004). Emotional effects of
startling background music during reading news reports: The
moderating influence of dispositional BIS and BAS sensitivities. Scandinavian Journal of Psychology, 45, 231-238.
*R AWLINGS , D., & L EOW , S. H. (2008). Investigating the role of
psychoticism and sensation seeking in predicting emotional
reactions to music. Psychology of Music, 36, 269-287.
*R ESNICOW , J. E., S ALOVEY , P., & R EPP , B. H. (2004). Is recognition of emotion in music performance an aspect of emotional intelligence? Music Perception, 22, 145-158.
*R ICKARD , N. S. (2004). Intense emotional responses to music: A
test of the physiological arousal hypothesis. Psychology of
Music, 32, 371-388.
*R ITOSSA , D. A., & R ICKARD , N. S. (2004). The relative utility of
pleasantness and liking dimensions in predicting the emotions expressed by music. Psychology of Music, 32, 5-22.
*R OBAZZA , C., M ACALUSO , C., & DU RSO , V. (1994). Emotional
reactions to music by gender, age, and expertise. Perceptual
and Motor Skills, 79, 939-944.
*R OY , M., M AILHOT , J.-P., G OSSELIN , N., P AQUETTE , S., &
P ERETZ , I. (2008). Modulation of the startle reflex by pleasant
and unpleasant music. International Journal of
Psychophysiology, 71, 37-42.
*R OY , M., P ERETZ , I., & R AINVILLE , P. (2008). Emotional
valence contributes to music-induced analgesia. Pain, 134,
140-147.
*R OZIN , A., R OZIN , P., & G OLDBERG , E. (2004). The feeling of
music past: How listeners remember musical affect. Music
Perception, 22, 15-39.
R USSELL , J. A. (1980). A circumplex model of affect. Journal of
Personality and Social Psychology, 39, 1161-1178.
R USSELL , J. A. (1983). Pancultural aspects of human conceptual
organization of emotions. Journal of Personality and Social
Psychology, 45, 1281-1288.
S ALOMONSSON , B. (1989). Music and affects: Psychoanalytic
viewpoints. Scandinavian Psychoanalytic Review, 12, 126-144.
*S AMMLER , D., G RIGUTSCH , M., F RITZ , T., & K OELSCH , S.
(2007). Music and emotion: Electrophysiological correlates of
the processing of pleasant and unpleasant music.
Psychophysiology, 44, 293-304.
*S A RKA MO , T., T ERVANIEMI , M., L AITINEN , S., F ORSBLOM , A.,
S OINILA , S., M IKKONEN , M., ET AL . (2008). Music listening
enhances cognitive recovery and mood after middle cerebral
artery stroke. Brain, 131, 866-876.
*S CHELLENBERG , E. G., K RYSCIAK , A. M., & C AMPBELL , R. J.
(2000). Perceiving emotion in melody: Interactive effects of
pitch and rhythm. Music Perception, 18, 155-171.
*S CHELLENBERG , E. G., N AKATA , T., H UNTER , P. G., & T AMOTO ,
S. (2007). Exposure to music and cognitive performance: tests
of children and adults. Psychology of Music, 35, 5-19.

This content downloaded from 129.240.0.83 on Mon, 6 May 2013 08:34:26 AM


All use subject to JSTOR Terms and Conditions

334

Tuomas Eerola & Jonna K. Vuoskoski

*S CHELLENBERG , E. G., P ERETZ , I., & V IEILLARD , S. (2008).


Liking for happy- and sad-sounding music: Effects of exposure. Cognition and Emotion, 22, 218-237.
*S CHERER , K. R. (1995). Expression of emotion in voice and
music. Journal of Voice, 9, 235-248.
*S CHERER , K. R. (2004). Which emotions can be induced by
music? What are the underlying mechanisms? And how can
we measure them? Journal of New Music Research, 33, 239-251.
SCHERER, K. R., BANSE, R., & WALLBOTT, H. G. (2001). Emotion
inferences from vocal expression correlate across languages
and cultures. Journal of Cross-Cultural Psychology, 32, 76-92.
S CHERER , K. R., & Z ENTNER , M. R. (2001). Emotional effects of
music: Production rules. In P. N. Juslin & J. A. Sloboda (Eds.),
Music and emotion: Theory and research (pp. 361-392).
Oxford, UK: Oxford University Press.
*S CHERER , K. R., Z ENTNER , M. R., & S CHACHT , A. (2002).
Emotional states generated by music: An exploratory study of
music experts. Musicae Scientiae, 6, 149-172.
SCHIMMACK, U., & GROB, A. (2000). Dimensional models of core
affect: A quantitative comparison by means of structural
equation modeling. European Journal of Personality, 14,
325-345.
*S CHMIDT , L. A., & T RAINOR , L. J. (2001). Frontal brain electrical activity (EEG) distinguishes valence and intensity of
musical emotions. Cognition and Emotion, 15, 487-500.
*S CHUBERT , E. (1996). Enjoyment of negative emotions in music:
An associative network explanation. Psychology of Music, 24,
18-28.
*S CHUBERT , E. (1999). Measuring emotion continuously:
Validity and reliability of the two-dimensional emotion-space.
Australian Journal of Psychology, 51, 154-165.
*S CHUBERT , E. (2001). Correlation analysis of continuous
emotional response to music: Correcting for the effects of serial
correlation. Musicae Scientiae, Special Issue, 213-236.
*S CHUBERT , E. (2003). Update of Hevners adjective checklist.
Perceptual and Motor Skills, 96, 1117-1122.
*S CHUBERT , E. (2004). Modeling perceived emotion with continuous musical features. Music Perception, 21, 561-585.
*S CHUBERT , E. (2007). The influence of emotion, locus of emotion and familiarity upon preference in music. Psychology of
Music, 35, 499-515.
*S CHUTZ , M., H URON , D., K EETON , K., & L OEWER , G. (2008).
The happy xylophone: acoustics affordances restrict an emotional palate. Empirical Musicology Review, 3, 126-135.
*S IEGWART , H., & S CHERER , K. (1995). Acoustic concomitants of
emotional expression in operatic singing: The case of Lucia in
Ardi gli incensi. Journal of Voice, 9, 249-260.
*S ILVIA , P. J., & A BELE , A. E. (2002). Can positive affect induce
self-focused attention? Methodological and measurement
issues. Cognition and Emotion, 16, 845-853.
*SLOBODA, J. A. (1991). Music structure and emotional response:
Some empirical findings. Psychology of Music, 19, 110-120.

*S LOBODA , J. A. (2001). The "sound of music" versus the "essence


of music": Dilemmas for music-emotion researchers:
Commentary. Musicae Scientiae, 6, 237-255.
*S LOBODA , J. A., & L EHMANN , A. C. (2001). Tracking performance correlates of changes in perceived intensity of emotion
during different interpretations of a Chopin piano prelude.
Music Perception, 19, 87-120.
S LOBODA , J. A., & ON EILL , S. A. (2001). Emotions in everyday
listening to music. In P. N. Juslin & J. A. Sloboda (Eds.), Music
and emotion: Theory and research (pp. 415-430). Oxford, UK:
Oxford University Press.
S OUSOU , S. D. (1997). Effects of melody and lyrics on mood and
memory. Perceptual and Motor Skills, 85, 31-40.
*S PACKMAN , M., F UJIKI , M., B RINTON , B., N ELSON , D., &
A LLEN , J. (2005). The ability of children with language
impairment to recognize emotion conveyed by facial
expression and music. Communication Disorders Quarterly,
26, 131-143.
*S PRECKELMEYER , K., K UTAS , M., U RBACH , T., A LTENMU LLER ,
E., & M U NTE , T. (2006). Combined perception of emotion in
pictures and musical sounds. Brain Research, 1070, 160-170.
*S TEINBEIS , N., K OELSCH , S., & S LOBODA , J. A. (2006). The role
of harmonic expectancy violations in musical emotions:
Evidence from subjective, physiological, and neural responses.
Journal of Cognitive Neuroscience, 18, 1380-1393.
S TRATTON , V. N., & Z ALANOWSKI , A. H. (1994). Affective
impact of music vs. lyrics. Empirical Studies of the Arts, 12,
173-184.
*S UDA , M., M ORIMOTO , K., O BATA , A., K OIZUMI , H., & M AKI ,
A. (2008). Emotional responses to music: Towards scientific
perspectives on music therapy. Neuroreport, 19, 75-78.
*S UNDBERG , J., I WARSSON , J., & H AGEGRD , H. (1994). A singers expression of emotions in sung performance. STL-QPSR,
35, 217-229.
*T AN , S., S PACKMAN , M., & B EZDEK , M. (2007). Viewers
interpretations of film characters emotions: Effects of presenting film music before or after a character is shown. Music
Perception, 25, 135-152.
*T ERWOGT , M. M., & V AN G RINSVEN , F. (1988). Recognition of
emotions in music by children and adults. Perceptual and
Motor Skills, 67, 697-698.
*T ERWOGT , M. M., & V AN G RINSVEN , F. (1991). Musical
expression of moodstates. Psychology of Music, 19, 99-109.
*T HAUT , M. H., & D AVIS , W. B. (1993). The influence of
subject-selected versus experimenter-chosen music on affect,
anxiety, and relaxation. Journal of Music Therapy, 30, 210-223.
T HAYER , R. E. (1989). The biopsychology of mood and arousal.
New York: Oxford University Press.
*T HAYER , J. F., & F AITH , M. L. (2001). A dynamic systems
model of musically induced emotions. Physiological and selfreport evidence. Annals of the New York Academy of Sciences,
930, 452-456.

This content downloaded from 129.240.0.83 on Mon, 6 May 2013 08:34:26 AM


All use subject to JSTOR Terms and Conditions

Review of Music and Emotion Studies

*T HOMA , M., R YF , S., E HLERT , U., & N ATER , U. (2006).


Regulation of emotions by listening to music in emotional
situations. In M. Baroni, A. R. Addessi, R. Caterina, & M. Costa
(Eds.), Proceedings of the 9th International Conference on
Music Perception and Cognition (pp. 1088-1093). Bologna,
Italy: ICMPC-ESCOM
*THOMPSON, W. F., & ROBITAILLE, B. (1992). Can composers express
emotions through music? Empirical Studies of the Arts, 10, 79-89.
*T HOMPSON , W., R USSO , F., & Q UINTO , L. (2008). Audio-visual
integration of emotional cues in song. Cognition and Emotion,
22, 14571470.
T HOMPSON , W. F., S CHELLENBERG , E. G., & H USAIN , G. (2001).
Arousal, mood, and the Mozart effect. Psychological Science,
12, 248-251.
*T IMMERS , R., & ASHLEY, R. (2007). Emotional ornamentation in
performances of a Handel sonata. Music Perception, 25, 117-134.
*T IMMERS , R., M AROLT , M., C AMURRI , A., & V OLPE , G. (2006).
Listeners emotional engagement with performances of
a Scriabin etude: An explorative case study. Psychology of
Music, 34, 481-510.
T REHUB , S. E., & N AKATA , T. (2002). Emotion and music in
infancy. Musicae Scientiae, Special issue 2001-2002, 3761.
T UGADE , M. M., F REDRICKSON , B. L., & F ELDMAN -B ARRETT , L.
(2004). Psychological resilience and positive emotional granularity: Examining the benefits of positive emotions on coping
and health. Journal of Personality, 72, 1161-1190.
*UNWIN, M. M., KENNY, D. T., & DAVIS, P. J. (2002). The effects
of group singing on mood. Psychology of Music, 30, 175-185.
V AITL , D., V EHRS , W., & S TERNAGEL , S. (1993). Prompts leitmotif - emotion: Play it again, Richard Wagner. In N.
hman (Eds.), The structure of emotion:
Birnbaumer & A. O
Psychophysiological, cognitive, and clinical aspects (pp.
169-189). Seattle, WA: Hogrefe & Huber.
*V ANDER A RK , S. D., & E LY , D. (1993). Cortisol, biochemical,
and galvanic skin responses to music stimuli of different
preference values by college students in biology and music.
Perceptual and Motor Skills, 77, 227-234.
*V A STFJA LL , D. (2002). Emotion induction through music: A
review of the musical mood induction procedure. Musicae
Scientiae, Special Issue, 173-209.
*V IEILLARD , S., P ERETZ , I., G OSSELIN , N., K HALFA , S., G AGNON ,
L., & B OUCHARD , B. (2008). Happy, sad, scary and peaceful
musical excerpts for research on emotions. Cognition and
Emotion, 22, 720-752.
*V INES , B., K RUMHANSL , C., W ANDERLEY , M., I OANA , M., &
L EVITIN , D. (2005). Dimensions of emotion in expressive
musical performance. Annals of New York Academy of
Sciences, 1060, 462-466.

335

*V INK , A. (2001). Music and emotion: Living apart together: A


relationship between music psychology and music therapy.
Nordic Journal of Music Therapy, 10, 144-158.
V UOSKOSKI , J. K., & E EROLA , T. (2011). The role of mood and
personality in the perception of emotions represented by
music. Cortex, 47, 1099-1106.
*W ATERMAN , M. (1996). Emotional responses to music: Implicit
and explicit effects in listeners and performers. Psychology of
Music, 24, 53-67.
W ATSON , D., C LARK , L. A., & T ELLEGEN , A. (1988).
Development and validation of brief measures of positive and
negative affect: The PANAS scales. Journal of Personality and
Social Psychology, 54, 1063-1070.
*W ATSON , D., W IESE , D., V AIDYA , J., & T ELLEGEN , A. (1999).
The two general activation systems of affect: Structural
findings, evolutionary considerations, and psychobiological
evidence. Journal of Personality and Social Psychology, 76,
820-838.
W EDIN , L. (1972). A multidimensional study of perceptualemotional qualities in music. Scandinavian Journal of
Psychology, 13, 241-257.
*W ELLS , A., & H AKANEN , E. (1991). The emotional use of
popular music by adolescents. Journalism Quarterly, 68,
445-454.
*W ITVLIET , C. V. O., & V RANA , S. R. (2007). Play it again Sam:
Repeated exposure to emotionally evocative music polarises
liking and smiling responses, and influences other affective
reports, facial EMG, and heart rate. Cognition and Emotion,
21, 1-23.
*W OODY , R. H. (2002). Emotion, imagery and metaphor in the
acquisition of musical performance skill. Music Education
Research, 4, 213-224.
*W U , T.-L., & J ENG , S.-K. (2006). Automatic emotion classification of musical segments. In M. Baroni, A. R. Addessi, R.
Caterina, & M. Costa (Eds.), Proceedings of the 9th
International Conference on Music Perception and Cognition
(pp. 385-393). Bologna, Italy: ICMPC-ESCOM.
*Y ANG , Y., L IN , Y., S U , Y., & C HEN , H. (2008). A regression
approach to music emotion recognition. IEEE Transactions on
Audio Speech and Language Processing, 16, 448-457.
*Z ANON , P., & P OLI , G. (2003). Estimation of parameters in rule
systems for expressive rendering of musical performance.
Computer Music Journal, 27, 29-46.
Z AJONC , R. B. (1980). Feeling and thinking: Preferences need no
inferences. American Psychologist, 35, 151-175.
*Z ENTNER , M., G RANDJEAN , D., & S CHERER , K. R. (2008).
Emotions evoked by the sound of music: Differentiation,
classification, and measurement. Emotion, 8, 494-521.

This content downloaded from 129.240.0.83 on Mon, 6 May 2013 08:34:26 AM


All use subject to JSTOR Terms and Conditions

336

Tuomas Eerola & Jonna K. Vuoskoski

Appendix
TABLE A1. Prevalence of Self-Report Approach to Music and Emotions Across Emotion Models (Columns) and Emotion Loci (Rows).

Discrete

Dimensional

Miscellaneous

Music-specific

Felt

Banich et al. (1992);


Baumgartner et al. (2006);
Etzel et al. (2006); Hunter
et al. (2008); Khalfa,
Delbe, et al. (2008);
Konecni et al. (2008);
Kreutz et al. (2002, 2008);
Krumhansl (1997);
Krumhansl, Wanderley,
Ioana, & Levitin (2005);
Mitterschiffthaler et al.
(2007); Nyklcek et al.
(1997); Roy et al. (2008b);
Scherer et al. (2002);
Thayer & Faith (2001);
Vines,

Alfredson, Risberg,
Hagberg, & Sustafson
(2004); Bigand et al.
(2005); Efimova & Budkya
(2006); Gabrielsson &
Lindstrom (2003); Gupta
& Gupta (2005); Husain
et al. (2002); Konecni,
Brown, & Wanic (2007);
Kreutz et al. (2008);
Leman et al. (2005); Miller
& Strongman (2002);
Matsuura (1998); Martin
& Metha (1997);
McAdams et al. (2004);
Nagel, Kopiez, Grewe, &
Altenmuller (2008);
Napoles & Madsen (2008);
Panksepp (1995); Rozin,
Rozin, & Goldberg (2004);
Schubert (2007); Sloboda
(1991); Thaut & Davis
(1993); Timmers, Marolt,
Camurri, & Volpe (2006)

Bartel (1992); Coffman


et al. (1995); Collier
(2007); Dibben (2004);
Gfeller et al. (1991);
Gfeller & Coffman (1991);
Giomo (1993); Juslin &
Laukka (2004); Juslin et al.
(2008); Kaminska &
Woolf (2000); Kinsella &
Jones (1990); Lindstrom
et al. (2003); Scherer et al.
(2002); Schubert (2003);
Wells & Hakanen (1991);
Zentner et al. (2008)

Perceived

Adachi & Trehub (2002);


Ali & Peynircioglu (2006);
Balkwill & Thompson
(1999); Balkwill,
Thompson, & Matsunaga
(2004); Behrens & Green
(1993); Boone &
Cunningham (2001);
Bouhuys, Bloem, &
Groothuis (1995);
Brosgole & Weisman
(1995); Bresin & Friberg
(2000); Chen, Yuan,
Huang, Chen, & Li (2008);
Collier & Hubbard (2001a,
2001b); Cunningham &
Sterling (1998); Dahl &
Friberg (2007); Dalla Bella

Altenmuller et al. (2002);


Blood & Zatorre (2001);
Blood et al. (1999); Collier
(2007); Dibben (2004);
Ellis & Simons (2005);
Evans & Schubert (2008);
Faith & Thayer (2001);
Gomez & Danuser (2004,
2007); Grewe et al.
(2007a); Hunter et al.
(2008); Husain et al.
(2002); Kabuto et al.
(1993); Kallinen & Ravaja
(2004, 2006); Khalfa,
Delbe, et al. (2008a);
Koelsch et al. (2006);
Koelsch, Fritz, & Schlaug
(2008); Konecni et al.
(2008); Kreutz, Bongard,
Rohrmann, Hodapp, &
Grebe (2004); Kreutz et al.
(2008); Krumhansl (1997);
Leman et al. (2005);
McFarland & Kennison
(1989a, 1989b);
Mitterschiffthaler et al.
(2007); Nater et al. (2006);
Nyklcek et al. (1997);
Ravaja & Kallinen (2004);
Rickard (2004); Roy,
Peretz, et al. (2008); Roy,
Mailhot, et al. (2008);
Schellenberg et al. (2007);
Schubert (2007); Thayer &
Faith (2001); Vines et al.
(2005); Webster & Weir
(2005); Witvliet & Vrana
(2007)
Costa, Fine, & Bitti (2004);
Dibben (2004); Eschrich,
Munte, & Altenmuller
(2008); Evans & Schubert
(2008); Gagnon & Peretz
(2000, 2003); Geringer
et al. (1996); Gorn, Pham,
& Sin (2001); Gosselin
et al. (2005, 2006, 2007);
Holbrook & Anand
(1992); Ilie & Thompson
(2006); Iwanaga &
Tsukamoto (1998);
Kallinen & Ravaja (2006);
Khalfa, Delbe, et al.
(2008); Khalfa et al.
(2002); Khalfa, Roy, et al.
(2008); Khalfa, Schon,

Baraldi, De Poli, & Roda`


(2006); Brittin & Duke
(1997); Costa, Bitti, &
Bonfiglioni (2000);
Fredrickson (1995); Frego
(1999); Geringer et al.
(1996); Juslin (1997c);
Kamenetsky et al. (1997);
Krumhansl (1998); Leman
et al. (2005); Lindstrom
(2006); Madison (2000);
Madsen & Frederickson
(1993); North &
Hargreaves (1997);
Rawlings & Leow (2008);
Ritossa & Rickard (2004);
Schubert (2003, 2007)

Bartel (1992); Gfeller et al.


(1991); Gfeller & Coffman
(1991); Coffman et al.
(1995); Dibben (2004);
Giomo (1993); Juslin &
Laukka (2004); Kaminska
& Woolf (2000); Schubert
(2003); Siegwart & Scherer
(1995); Zentner et al.
(2008)

(continued)

This content downloaded from 129.240.0.83 on Mon, 6 May 2013 08:34:26 AM


All use subject to JSTOR Terms and Conditions

Review of Music and Emotion Studies

337

Table A1. (continued)

Discrete

Dimensional

et al. (2001); Fritz et al.


(1999); Gabrielsson &
Juslin (1996); Gabrielsson
& Lindstrom (1995);
Gerardi & Gerken (1995);
Gilboa, Bodner, & Amir
(2006); Gloeckner (2006);
Gosselin et al. (2005, 2006,
2007); Gregory & Varney
(1996); Green et al. (2008);
Juslin (1997a, 1997b,
1997c, 2000); Juslin &
Madison (1999); Kallinen
(2005); Kastner &
Crowder (1990); Juslin
Meyer, Palmer, & Mazo
(1998); Khalfa et al.
(2002); Khalfa, Delbe,
et al. (2008); Khalfa, Roy,
et al. (2008); Konecni et al.
(2008); Lindstrom et al.
(2006); Lu, Liu, & Zhang
(2006); Madison (2000);
Mancini, Bresin, &
Pelachaud (2007);
Morozov (1996); Nawrot
(2003); Morton & Trehub
(2007); North &
Hargreaves (1997);
Resnicow et al. (2004);
Robazza et al. (1994); Roy,
Mailhot, et al. (2008);
Sarkamo et al. (2008);
Schellenberg, Krysciak, &
Campbell (2000);
Schellenberg et al. (2008);
Spackman, Fujiki,
Brinton, Nelson, & Allen
(2005); Spreckelmeyer,
Kutas, Urbach,
Altenmuller, & Munte
(2006); Tan et al. (2007);
Terwogt & Van Grinsven
(1988, 1991); Timmers &
Ashley (2007); Vieillard
et al. (2008)
Felt
Khalfa, Delbe, et al.
Perceived (2008); Konecni et al.
(2008)

Miscellaneous

Music-specific

Leman et al. (2005);


Panksepp (1995);
Schubert (2007)

Dibben (2004); Juslin &


Laukka (2004); Zentner
et al. (2008)

Anton, & LiegeoisChauvel (2005); Koelsch


et al. (2008a); Kopiez
(2007); Konecni et al.
(2008); Korhonen et al.
(2006); Leman et al.
(2005); Lindstrom (2003);
Lindstrom (2006); Luck
et al. (2008); MacDorman,
Ough, & Ho (2007);
Madsen (1997a, 1998);
Pallesen et al. (2005);
Rawlings & Leow (2008);
Ritossa & Rickard (2004);
Roy, Mailhot, et al.
(2008b); Schubert (1999,
2003, 2004, 2007);
Spreckelmeyer et al.
(2006); Terwogt &
Van Grinsven (1991);
Thoma, Ryf, Ehlert, &
Nater (2006); Thompson,
Russo, & Quinto (2008);
Vieillard et al. (2008); Wu
& Jeng (2006); Yang, Lin,
Su, & Chen (2008)

Dibben (2004); Evans &


Schubert (2008); Kallinen
& Ravaja (2006); Khalfa,
Delbe, et al. (2008);
Konecni et al. (2008);
Leman et al. (2005);
Schubert (2007)

(continued)

This content downloaded from 129.240.0.83 on Mon, 6 May 2013 08:34:26 AM


All use subject to JSTOR Terms and Conditions

338

Tuomas Eerola & Jonna K. Vuoskoski

Table A1. (continued)

N/A

Discrete

Dimensional

Miscellaneous

Music-specific

Erkkila et al. (2008);


Lundqvist, Carlsson,
Hilmersson & Juslin
(2009); Pigniatello &
Rasar (1989); Thompson
& Robitaille (1992)

Erkkila et al. (2008);


Schmidt & Trainor (2001);
Schubert (2001); Unwin,
Kenny, & Davis (2002)

Grewe et al. (2007b); Herz Kinsella & Jones (1990);


Lindstrom et al. (2003);
(1998); Hoshino (1996);
Lehmann (1997); Lowis
(2002); Madsen (1997b);
Silvia & Abele (2002);
Sloboda & Lehmann
(2001); Steinbeis, Koelsch,
& Sloboda (2006);
Waterman (1996); Woody
(2002)

TABLE A2. Prevalence of Music Analytic Approach to Music and Emotions Across Emotion Models (Columns) and Emotion Loci (Rows).

Discrete
Felt

Perceived

Dimensional

Other

Dibben (2006); Grewe et al.


(2007a); Husain et al. (2002);
Leman et al. (2005); McAdams
et al. (2004); Nagel et al.
(2008); Sloboda (1991);
Steinbeis et al. (2006);
Timmers et al. (2006)
Costa et al. (2004); Gagnon & Kamenetsky (1997);
Adachi & Trehub (1998);
Krumhansl (1998); Leman
Balkwill & Thompson (1999); Peretz (2003); Ilie &
Balkwill et al. (2004); Bresin & Thompson (2006); Korhonen et al. (2005)
et al.; (2006); Leman et al.
Friberg (2000); Dalla Bella
et al. (2001); Fritz et al. (2009); (2005); Livingstone,
Muhlberger, Brown, & Loch;
Gabrielsson & Juslin (1996);
(2007); Luck et al. (2006,
Gabrielsson & Lindstrom
2008); MacDorman et al.
(1995); Juslin (1997a, 1997b,
2000); Juslin & Laukka (2000); (2007); Schubert (2004);
Vieillard et al.; (2008); Wu &
Juslin & Laukka (2003);
Jeng (2006); Yang et al. (2008)
Laukka (2007); Laukka &
Gabrielsson (2000); Lu et al.
(2006); Morozov et al. (1996);
Schellenberg et al. (2000);
Sundberg, Iwarsson, &
Hagegrd (1994); Thompson
& Robitaille (1992); Timmers
& Ashley (2007); Vieillard
et al. (2008);
Leman et al. (2005)
Leman et al. (2005)
Hunter et al. (2008)

Felt
Perceived
N/A
Goydke et al. (2004); Huron
et al. (2008); Schutz, Huron,
Keeton, & Loewer (2008)

Grewe et al. (2007a); Hunter


et al. (2008); Husain et al.
(2002); Leman et al. (2005)

Ockelford (2005); Sloboda


(2001); Waterman (1996);
Zanon & Poli (2003);

This content downloaded from 129.240.0.83 on Mon, 6 May 2013 08:34:26 AM


All use subject to JSTOR Terms and Conditions

Music-specific
-

Kaminska &
Woolf (2000);
Siegwart &
Scherer
(1995)

Review of Music and Emotion Studies

339

TABLE A3. Prevalence of Biological Approach to Music and Emotions Across Emotion Models (Columns) and Emotion Loci (Rows).

Discrete

Dimensional

Other

Music-specific

Felt

Banich et al. (1992);


Baumgartner et al. (2006);
Etzel et al. (2006); Kreutz et al.
(2002); Krumhansl (1997);
Mitterschiffthaler et al. (2007);
Nyklcek et al. (1997); Suda,
Morimoto, Obata, Koizumi, &
Maki (2008)

Alfredson et al. (2004);


Chapados & Levitin (2008);
Grewe et al. (2007a, 2007b);
Gupta & Gupta (2005);
Krumhansl (1997); Matsuura
(1998); VanderArk & Ely
(1993); Steinbeis et al. (2006)

Dibben (2004)

Perceived

Gloeckner (2006); Gosselin


et al. (2005, 2006, 2007);
Khalfa et al. (2002); Khalfa,
Roy, et al. (2008); Peretz &
Gagnon (1999); Peretz et al.
(1998, 2001); Spreckelmeyer
et al. (2006)

Altenmuller et al. (2002);


Blood & Zatorre (2001); Blood
et al. (1999); Dibben (2004);
Ellis & Simon (2005); FloresGutierrez et al. (2007); Gomez
& Danuser (2007); Grewe et al.
(2007a); Kabuto et al. (1993);
Kallinen & Ravaja (2004); Kim
& Andre (2008); Koelsch et al.
(2006); Kreutz et al. (2004);
Krumhansl (1997); McFarland
& Kennison (1989a, 1989b);
Mitterschiffthaler et al. (2007);
Nater et al. (2006); Nyklcek
et al. (1997); Ravaja & Kallinen
(2004); Rickard (2004); Roy
et al. (2008a, 2008b); Sammler
et al. (2007); Schmidt &
Trainor (2001); Witvliet &
Vrana (2007);
Brown et al. (2004); Dibben
(2004); Gagnon & Peretz
(2000); Gosselin et al. (2005,
2006, 2007); Khalfa et al.
(2002, 2005); Khalfa, Roy,
et al. (2008) Pallesen et al.
(2005); Peretz et al. (2001);
Van Strien & Boon (1997);
Spreckelmeyer et al. (2006)
Dibben (2004)

Brown et al. (2004); Peretz


et al. (2001)

Dibben (2004)

Dibben (2004)

Erkkila et al. (2008)

Herz (1998); Iwaki, Hayashi, & Hori (1997); Steinbeis et al.


(2006)

Felt
Perceived
N/A
Erkkila et al. (2008); Goydke
et al. (2004); Lundqvist et al.
(2009); Panksepp & Bekkedal
(1997); Pigniatello & Rasar
(1989)

TABLE A4. Prevalence of Clinical Approach to Music and Emotions Across Emotion Models (Columns) and Emotion Loci (Rows).

Felt
Perceived

Discrete

Dimensional

Other

Khalfa, Delbe, et al. (2008)


Dellacherie et al. (2008); Gosselin et al.
(2005, 2006, 2007); Khalfa, Delbe, et al.
(2008); Peretz et al. (1998, 2001);
Sarkamo et al. (2008); Spackman et al.
(2005)
Khalfa, Delbe, et al. (2008)

Khalfa, Delbe, et al. (2008)


Altman et al. (2000); Gosselin et al.
(2005, 2006, 2007); Khalfa, Delbe, et al.
(2008); Luck et al. (2006); Peretz et al.
(2001)

Nielzen et al. (1993);


Peretz et al.
(2001)

Khalfa, Delbe, et al. (2008)

Erkkila et al. (2008)

Kinsella &
Jones (1990)

Felt
Perceived
N/A
Erkkila et al. (2008)

This content downloaded from 129.240.0.83 on Mon, 6 May 2013 08:34:26 AM


All use subject to JSTOR Terms and Conditions

Music-specific

340

Tuomas Eerola & Jonna K. Vuoskoski

TABLE A5. Prevalence of Developmental Approach to Music and Emotions Across Emotion Models (Columns) and Emotion Loci (Rows).

Felt
Perceived

Discrete

Dimensional

Other

Music-specific

Adachi & Trehub (1998, 2002); Boone & Cunningham


(2001); Brosgole & Weisman (1995); Cunningham & Sterling
(1988); Dalla Bella et al. (2001); Flom, Gentile, & Pick (2008);
Gerardi & Gerken (1995); Kastner & Crowder (1990);
Morton & Trehub (2007); Nawrot (2003); Terwogt &
Van Grinsven (1991)
-

Terwogt &
Van Grinsven
(1991)

Boone &
Cunningham
(2001)

Giomo (1993)

Felt
Perceived
N/A
-

TABLE A6. Prevalence of Individual and Cross-Cultural Approach to Music and Emotions Across Emotion Models (Columns) and Emotion Loci
(Rows).

Discrete

Dimensional

Other

Music-specific

Felt

Kreutz, Bongard, & Von Jussis


(2002); Kreutz, Ott, et al. (2008)

Balkwill & Thompson (1999);


Balkwill et al. (2004); Fritz et al.
(2009); Gregory & Varney (1996);
Meyer et al. (1998);
-

Bigand et al. (2005);


Efimova & Budyka
(2006); Kreutz, Ott,
et al. (2008)
Rawlings & Leow
(2008)

Perceived

Chen, Zhou, & Bryant (2007);


Kabuto et al. (1993); Kallinen &
Ravaja (2004); Kreutz, Ott, et al.
(2008); Nater et al. (2006)
Rawlings & Leow (2008)

Felt
Perceived

TABLE A7. List of Studies Applying Mainly Theoretical Approach to Music and Emotions.

Author
Bever (1988); Cook (2007); Dibben (2006; Gabrielsson (2002); Hargreaves & North (1999); Holbrook & Anand (1992); Juslin
et al. (2001); Juslin & Laukka (2003, 2004); Juslin & Vastfjall (2008);
Juslin & Zentner (2001); Koelsch (2005); Konecni (2008); Konecni et al. (2008); Livingstone et al. (2007); London (2001);
Margulis (2008); McEvilly (1999); Ockelford (2005); Panksepp (1992, 1995); Salomonsson (1989); Scherer (1995, 2004); Schubert
(1996); Sloboda (2001); Vastfjall (2002); Vink (2001)

This content downloaded from 129.240.0.83 on Mon, 6 May 2013 08:34:26 AM


All use subject to JSTOR Terms and Conditions